Download Last millennium Northern Hemisphere summer temperatures from

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Fred Singer wikipedia , lookup

Early 2014 North American cold wave wikipedia , lookup

Global warming controversy wikipedia , lookup

Climate governance wikipedia , lookup

Citizens' Climate Lobby wikipedia , lookup

Climatic Research Unit email controversy wikipedia , lookup

Climate engineering wikipedia , lookup

Global warming wikipedia , lookup

Effects of global warming on human health wikipedia , lookup

Media coverage of global warming wikipedia , lookup

Climate change and agriculture wikipedia , lookup

Climate change in Tuvalu wikipedia , lookup

Climate change feedback wikipedia , lookup

Public opinion on global warming wikipedia , lookup

Climate change in the United States wikipedia , lookup

Scientific opinion on climate change wikipedia , lookup

Climate change and poverty wikipedia , lookup

Effects of global warming on humans wikipedia , lookup

Soon and Baliunas controversy wikipedia , lookup

Global warming hiatus wikipedia , lookup

General circulation model wikipedia , lookup

Wegman Report wikipedia , lookup

Solar radiation management wikipedia , lookup

Years of Living Dangerously wikipedia , lookup

Attribution of recent climate change wikipedia , lookup

IPCC Fourth Assessment Report wikipedia , lookup

Climate sensitivity wikipedia , lookup

Climate change, industry and society wikipedia , lookup

Surveys of scientists' views on climate change wikipedia , lookup

Michael E. Mann wikipedia , lookup

Climatic Research Unit documents wikipedia , lookup

Hockey stick controversy wikipedia , lookup

Instrumental temperature record wikipedia , lookup

North Report wikipedia , lookup

Transcript
Quaternary Science Reviews 163 (2017) 1e22
Contents lists available at ScienceDirect
Quaternary Science Reviews
journal homepage: www.elsevier.com/locate/quascirev
Invited review
Last millennium Northern Hemisphere summer temperatures from
tree rings: Part II, spatially resolved reconstructions
Kevin J. Anchukaitis a, b, c, *, Rob Wilson d, c, Keith R. Briffa e, Ulf Büntgen f, g, h,
Edward R. Cook c, Rosanne D'Arrigo c, Nicole Davi c, i, Jan Esper j, David Frank b, g,
€ rn E. Gunnarson k, Gabi Hegerl l, Samuli Helama m, Stefan Klesse b, g, Paul J. Krusic k, c, n,
Bjo
Hans W. Linderholm o, Vladimir Myglan p, Timothy J. Osborn e, Peng Zhang o,
Milos Rydval d, q, Lea Schneider r, Andrew Schurer l, Greg Wiles s, Eduardo Zorita t
a
School of Geography and Development, University of Arizona, Tucson, AZ, USA
Laboratory of Tree-Ring Research, University of Arizona, Tucson, AZ, USA
c
Lamont-Doherty Earth Observatory of Columbia University, Palisades, NY, USA
d
School of Geography and Geosciences, University of St Andrews, UK
e
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich, UK
f
Department of Geography, University of Cambridge, Cambridge, UK
g
Swiss Federal Research Institute WSL, Birmensdorf, Switzerland
h
Global Change Research Centre and Masaryk University Brno, Czechia
i
Department of Environmental Science, William Paterson University, Wayne, NJ, USA
j
Department of Geography, Gutenberg University, Mainz, Germany
k
Department of Physical Geography, Stockholm University, Stockholm, Sweden
l
School of GeoSciences, University of Edinburgh, Edinburgh, UK
m
Natural Resources Institute Finland, Rovaniemi, Finland
n
Navarino Environmental Obs., Messinia, Greece
o
€teborg, Sweden
Department of Earth Sciences, University of Gothenburg, Go
p
Siberian Federal University, Krasnoyarsk, Russia
q
Faculty of Forestry and Wood Sciences, Czech University of Life Sciences Prague, Czechia
r
Department of Geography, Justus Liebig University, Giessen, Germany
s
Tree Ring Lab, The College of Wooster, Wooster, OH, USA
t
Institute of Coastal Research, Helmholtz-Zentrum Geesthacht (HZG), Hamburg, Germany
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 August 2016
Received in revised form
20 February 2017
Accepted 21 February 2017
Climate field reconstructions from networks of tree-ring proxy data can be used to characterize regionalscale climate changes, reveal spatial anomaly patterns associated with atmospheric circulation changes,
radiative forcing, and large-scale modes of ocean-atmosphere variability, and provide spatiotemporal
targets for climate model comparison and evaluation. Here we use a multiproxy network of tree-ring
chronologies to reconstruct spatially resolved warm season (MayeAugust) mean temperatures across
the extratropical Northern Hemisphere (40-90 N) using Point-by-Point Regression (PPR). The resulting
annual maps of temperature anomalies (750e1988 CE) reveal a consistent imprint of volcanism, with
96% of reconstructed grid points experiencing colder conditions following eruptions. Solar influences are
detected at the bicentennial (de Vries) frequency, although at other time scales the influence of insolation variability is weak. Approximately 90% of reconstructed grid points show warmer temperatures
during the Medieval Climate Anomaly when compared to the Little Ice Age, although the magnitude
varies spatially across the hemisphere. Estimates of field reconstruction skill through time and over space
can guide future temporal extension and spatial expansion of the proxy network.
© 2017 Elsevier Ltd. All rights reserved.
Keywords:
Tree-rings
Northern Hemisphere
Last millennium
Common Era
Summer temperatures
Reconstruction
Spatial
* Corresponding author. School of Geography and Development, University of
Arizona, Tucson, AZ, USA.
E-mail address: [email protected] (K.J. Anchukaitis).
http://dx.doi.org/10.1016/j.quascirev.2017.02.020
0277-3791/© 2017 Elsevier Ltd. All rights reserved.
1. Introduction
Global and hemispheric temperature anomalies reflect the
2
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
influence of both internal variability in the climate system as well
as the consequences of changes in radiative forcing, such as insolation, volcanic eruptions, and greenhouse gas concentrations.
Surface temperature is determined by the planetary energy balance
and serves as a symptom of perturbations to that balance, but also
contains variability due to natural climate system dynamics. Rising
global mean surface temperature is a key diagnostic for the influence of increasing greenhouse gases on the Earth's climate system.
Yet changes in incoming solar radiation, orbital (Milankovich)
changes, albedo and land use alterations, and natural and anthropogenic aerosols also influence surface temperatures. Different
radiative forcing mechanisms as well as internal modes of coupled
ocean-atmosphere variability may have distinct fingerprints on
temperature anomalies across different spatial, temporal, and
seasonal scales (Hegerl et al., 1997; Rind et al., 1999; Shindell et al.,
2001b; Hegerl et al., 2003; Rind et al., 2004; Shindell et al., 2003,
2004; Hegerl et al., 2006, 2007; Shindell and Faluvegi, 2009;
Shindell, 2014; Shindell et al., 2015). Surface temperature anomalies are therefore controlled by the superposition of various
external radiative and internal dynamical influences on the climate
system. Detection and attribution of the causes of temperature
fluctuations, as well as the prediction of future regional-scale
changes, thus depend on accurate quantification and understanding of spatial and temporal variations in surface temperature
(Hegerl et al., 1997; Stott and Tett, 1998; Meehl et al., 2004; Lean
and Rind, 2008; Stott and Jones, 2009; Stott et al., 2010; Solomon
et al., 2011; Hegerl and Stott, 2014).
Paleoclimate reconstructions of past temperature extend
knowledge of climate system variability beyond that available from
the limited instrumental observational record. They offer longer
timescales over which to observe a more complete range of variability in solar and volcanic forcing, extended opportunities to
characterize internal climate system fluctuations at decadal and
longer timescales, and the potential to separate forced and unforced responses to better understand their magnitude and
spatiotemporal patterns (Hegerl et al., 2003, 2007). Spatiallyexplicit reconstructions provide additional opportunities to refine
our understanding of fundamental climate system characteristics,
diagnose the influence of different forcings on various aspects of
the climate system, and provide insight into both regional climate
changes and the response of large-scale modes of oceanatmosphere variability (Seager et al., 2007; Cook et al., 2010a, b;
Hegerl and Russon, 2011; Phipps et al., 2013; PAGES 2k-PMIP3
group, 2015; Goosse, 2016). Comparisons between paleoclimatic
data and models also provide out-of-sample tests of the general
circulation models (GCMs) used for future climate projections and
can indicate where the modeled forced response or internal variability requires further evaluation and continued refinement. Such
comparisons may help constrain the probable range of model parameters or identify the forcing configurations most consistent
with past climate variability (Edwards et al., 2007; Anchukaitis
et al., 2010; Schmidt, 2010; Hegerl and Russon, 2011; Brohan
et al., 2012; Schurer et al., 2013; Schmidt et al., 2014; Harrison
et al., 2015; Tingley et al., 2015).
Reconstructions of last millennium and Common Era surface
temperatures have focused predominantly on single time-series to
represent continental- to global-scale variations in mean annual or
growing season temperatures aggregated over space (Frank et al.,
2010; Masson-Delmotte et al., 2013; PAGES2k, 2013; Stoffel et al.,
2015; Smerdon and Pollack, 2016) while fewer have used climate
field reconstruction (CFR) methods (Fritts, 1991; Cook et al., 1994;
Evans et al., 2001; Tingley et al., 2012) to quantify past temperature anomalies simultaneously through time and across space (c.f.
Mann et al., 1998; Tingley and Huybers, 2013; Wang et al., 2015).
Spatial field reconstructions offer the benefit of characterizing
regional-scale climate changes, can reveal spatial anomaly patterns
or fingerprints associated with atmospheric circulation, radiative
forcing, and large-scale modes of ocean-atmosphere variability, and
provide complete spatiotemporal targets for GCM evaluation
(Evans et al., 2001; Anchukaitis and McKay, 2014; Kaufman, 2014;
Schmidt et al., 2014).
Here, we develop and evaluate a climate field reconstruction of
extratropical Northern Hemisphere summer temperatures using an
updated network of temperature-sensitive tree-ring proxy chronologies and existing temperature reconstructions back to 750 CE
(Wilson et al., 2016). We are motivated by two fundamental challenges to the development of skillful large-scale last millennium
temperature reconstructions revealed over the last two decades
(c.f. Frank et al., 2010; Smerdon and Pollack, 2016): First, biases
arising from characteristics of the proxies themselves; second,
uncertainties arising from the choice of reconstruction
methodologies.
Tree-ring proxies provide precise annual dating and are broadly
distributed across extra-tropical land areas, making them one of
the most widely used proxies for climate reconstructions of the
Common Era (Hughes, 2002; Jones et al., 2009; Smerdon and
Pollack, 2016). Yet despite these advantages, certain challenges or
limitations exist: they preferentially reflect growing season temperature conditions, they require some manner of processing to
remove non-climatic age or tree geometry related growth trends,
and there exist a wide range of climate responses amongst the
more than two thousand tree-ring chronologies currently archived
in public repositories (Briffa, 1995, 2000; Briffa et al., 2002, 2004; St.
George, 2014; St. George and Ault, 2014). A decade ago, D'Arrigo
et al. (2006) and Wilson et al. (2007) used small high-latitude
networks of tree-ring proxy chronologies to reconstruct mean
annual Northern Hemisphere temperatures. These and subsequent
efforts have illuminated several extant challenges: a relatively
limited number of unambiguously temperature-sensitive chronologies, a predominance of ring-width chronologies in comparison to
the more temperature-sensitive wood density measurements
(D'Arrigo et al., 1992; Schweingruber et al., 1993; Briffa et al., 2004;
Frank et al., 2007; D'Arrigo et al., 2009; Esper et al., 2015; Wilson
et al., 2016), and the influence of non-stationarity in climate/tree
growth associations (‘divergence’; Briffa et al., 1998b; Wilson et al.,
2007; D'Arrigo et al., 2008), a particular problem for North American treeline tree-ring width chronologies (Jacoby and D'Arrigo,
1995; Andreu-Hayles et al., 2011; Anchukaitis et al., 2013) and
many previously collected wood density chronologies (Briffa et al.,
2002). Since the publication of D'Arrigo et al. (2006) and Wilson
et al. (2007), dozens of new tree-ring chronologies and local temperature reconstructions have become available, including many
new and updated latewood density (MXD) measurement series
that do not appear to exhibit any divergence (c.f. D'Arrigo et al.,
2009; Esper et al., 2010; Anchukaitis et al., 2013). We draw on
these new, published, and updated data here to develop a spatial
reconstruction of past summer temperature stretching back to 750
CE. This work extends the non-spatial hemisphere mean reconstruction published by Wilson et al. (2016).
Methodologies for last millennium climate reconstructions have
been extensively investigated and tested over the last two decades
(Mann and Rutherford, 2002; Rutherford et al., 2003; Zorita et al.,
2003; von Storch et al., 2004; Esper et al., 2005; Mann et al.,
2005, 2007; Li et al., 2010; Lee et al., 2008; Smerdon et al., 2008,
2010, 2011; Wang et al., 2015; Smerdon and Pollack, 2016). However, neither reduced space, empirical orthogonal regression
methods (c.f. Fritts, 1991; Cook et al., 1994; Mann et al., 1998) nor
most variants of regularized expectation maximization (RegEM
Schneider, 2001; Rutherford et al., 2003; Mann et al., 2009)
explicitly consider the location of the proxies relative to the
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
reconstruction target, and therefore reconstructions of individual
grid points can be significantly influenced by distant proxy sites.
While there is potential value in taking into account large-scale
teleconnections between the climate in a remote region and the
local conditions controlling proxy formation (Evans et al., 2001,
2002), the potential disadvantage to these approaches is that they
rely implicitly on the existence, stability, and persistence of those
teleconnections (Gershunov and Barnett, 1998; Rimbu et al., 2003;
Anchukaitis et al., 2006; Wilson et al., 2010; Lehner et al., 2012;
Gallant et al., 2013; Ortega et al., 2015; Wise, 2015; Lewis and
LeGrande, 2015), assume the time-stability of large-scale covariance patterns, and risk admitting distal predictors with spurious
relationships to local temperatures. However, several methods
exist which do account for the spatial distribution and relationship
of the proxy predictor network to the target field (Cook et al., 1999;
Tingley and Huybers, 2010; Steiger et al., 2014). Here we use Pointby-Point regression (PPR; Cook et al., 1999) to develop, for the first
time, a hemisphere-scale spatial temperature reconstruction. For
an extensive review of the full range of climate field reconstruction
methods, see Tingley et al. (2012).
We seek a spatial reconstruction of past temperatures that utilizes a predictor network with a clear biophysical and statistical
relationship with temperature. We therefore apply the expert
knowledge of the original developers of the individual chronologies
and reconstructions, relying on their experience with respect to
climate signal evaluation and statistical treatment (c.f. Esper et al.,
2016). PPR is a well-established, transparent, and spatially-explicit
methodology for climate field reconstruction. The result of PPR
here is a gridded map of summer temperature anomalies for each
year of the last ~1200 years with associated validation skill metrics
in both space and time. An additional benefit of PPR is that limiting
the grid point reconstructions to proximal predictors and avoiding
assumptions about the global covariance structure ensures that
distant grid points remain independent of one another. The spatial
features of the field can therefore be examined and compared
without concern that they arise from use of the same predictors.
We compare our results against radiative forcing over the last
millennium (Schmidt et al., 2012), diagnosing the potential role of
volcanic eruptions and insolation variability in shaping the
Northern Hemisphere extratropical warm season temperature
response across space and time. Our field reconstruction can serve
as a resource for understanding temperature variability in the past,
for comparison with other proxy records of environmental and
climate change, and to provide context for coupled human-natural
systems response to climate variability (e.g. Buckley et al., 2010;
Büntgen et al., 2011, 2016; Pederson et al., 2014). Our reconstruction also provides a spatiotemporal target appropriate for formal
detection and attribution of the influence of different sources of
radiative forcing on the Earth's climate system.
2. Materials and methods
2.1. Tree-ring series
Tree-ring based reconstructions of large-scale climate variability, whether a single mean time series or a spatial climate field
reconstruction, require selecting potential proxy predictors from
the many thousands of chronologies that have been developed over
the last century of tree-ring research (St. George, 2014). The majority of these chronologies were not collected or developed with
the intended purpose of temperature reconstruction and do not
contain a primary temperature signal. It is therefore necessary to
apply some selection procedure, lest chronologies lacking temperature information overwhelm the predictor set with ‘noise’
associated with soil moisture, archaeological selection, ecological
3
processes, or spurious non-temperature signals. Two broad categories of approach have been used for initial predictor selection:
statistical screening and expert assessment. In statistical screening
(c.f. Mann et al., 2008), proxy series are assessed for their correlations against local or regional observational temperature data, with
those data passing some significance, sign, or effect size threshold
then admitted to the pool of potential predictors. While this
approach clearly has merit as an objective and automated
approach, it virtually guarantees the selection of a proportion of
proxies without a realistic biophysical or substantial statistical association with the climate variable targeted for reconstruction. A
second approach utilizes expert assessment of individual chronologies based on an understanding of whether a proxy has the
requisite ecological, biological, geographical, and climatological
characteristics to serve as a reasonable temperature proxy.
Although this approach is not completely isolated from statistical
considerations, the advantage is that it strongly reduces the likelihood that a non-temperature proxy or nonsense predictor will
enter into a temperature reconstruction model. Expert assessment
can include not only climate signal and ecological criteria, but also
the methods used to develop the proxy series (e.g. Esper et al.,
2016). A potential disadvantage to this approach is that it is
partially subjective and therefore different investigators could
make different selections for the predictor pool. Hybrid approaches
that combine simple mechanistic or physiological assessment with
statistical evaluation have also been applied (c.f. Tierney et al.,
2015).
Here, we follow Wilson et al. (2016) and use as our predictor
series only published tree-ring chronologies and temperature reconstructions that demonstrate an established and biophysically
reasonable association with local temperatures. These include both
tree-ring chronologies as well as existing temperature reconstructions from Northern Hemisphere high-latitude and highaltitude locations, where dendrochronological and ecological
principles suggest the most limiting factor for growth is temperature (Fritts, 1976). We exclude from consideration chronologies
south of 40 N to avoid confounding climate signals associated with
moisture-sensitive trees (St. George, 2014) and we also reject
chronologies known to demonstrate evidence of the ‘divergence
problem’ (Briffa et al., 1998b; D'Arrigo et al., 2008, 2009; Wilson
et al., 2007), a problem previously observed to affect North American Picea glauca tree-ring width chronologies (D'Arrigo et al., 1992,
2009; Andreu-Hayles et al., 2011; Esper et al., 2012) and many MXD
chronologies developed in the 1970s and 1980s. We require that
our predictors extend back to at least 1750 CE and completely
forward to 1988 CE. The predictors retain the detrending and
standardization choices of the original authors. The resulting
dataset is designated N-TREND2015 and is composed of a mixture
of tree-ring width (TRW), MXD (Schweingruber et al., 1978), and
blue intensity (BI; McCarroll et al., 2002) data. N-TREND2015 is
archived and publicly available1 and is the same dataset used in
Part 1 of this study (Wilson et al., 2016). N-TREND is intended to be
a ‘living dataset’ that will grow or be modified as new proxies
become available or are updated. Details of the predictor time series used here and in Wilson et al. (2016) are available in Table 1. As
shown in Fig. 1, the NTREND-2015 network reflects a mix of proxy
types, dominated by MXD or BI (43 series, vs. 11 composed of treering width data only). A total of 54 series are available from 1750 to
1988 CE, the time period of full (denoted ‘BEST’) coverage of the
network. The number of sites drops precipitously toward the present, down to 34 by 1990 CE and to 25 series by 2000 CE, with only
3 sites remaining by 2011. There are 23 series at 1000 CE and 4
1
https://www.ncdc.noaa.gov/paleo[HYPHEN]search/study/19743.
4
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Table 1
Tree-ring chronology and temperature reconstruction predictor series used in the reconstruction. Additional information is available in Wilson et al. (2016). Latitude is given in
degrees North, and longitude is in degrees East. A range of latitude/longitude indicates the data or reconstruction cover a larger region. Detrending: STD: traditional detrending
standardization; RCS: regional curve standardization; SF: signal free extension to STD or RCS standardization. y indicates series used in D'Arrigo et al. (2006) and z those used in
Wilson et al. (2007).
Site Name
Code
Latitude
Longitude
Time Span
Proxy Type Detrending Citation
North America
Seward
NTR
MXD
STD
D'Arrigo et al., 2004
GOA
800e2010
TRW
RCS
Wiles et al., 2014
Wrangells
WRAx
1593e1992
MXD
STD
Davi et al., 2003z
Firth
Southern Yukon
Northern Yukon
Int. British Columbia
FIRT
YUS
YUN
IBC
162.18
to 162.27
149.31
to 141.42
145.00
to 140.00
141.38
140 to 133
125 to 135
121.43
to 117.03
117.19
110 to 120
115.55
103.52
76.00
74.00
70 to 65
62.25 to 61.56
1710e2001
Coastal Alaska
65.11 to
65.22
60.01 to
60.45
60e65
1073e2002
1684e2000
1638e1988
1600e1995
RCS-SF
STD
STD
STD/SF
Anchukaitis et al., 2013
Youngblut and Luckman, 2008z
Szeicz and MacDonald, 1995y
Wilson et al., 2014
918e1994
1135e1992
1551e2003
1492e2004
1373e1988
910e2011
1642e2002
1710e1998
MXD
TRW
TRW
TRW/MXD/
BI
RW/MXD
TRW
MXD
MXD
MXD
TRW
TRW
TRW/MXD
RCS
STD
STD
STD
RCS
RCS
STD
STD/RCS
Luckman and Wilson, 2005y
Biondi et al., 1999z
D'Arrigo et al., 2009, Anchukaitis et al., 2013
D'Arrigo et al., 2009, Anchukaitis et al., 2013
Schneider et al., 2015
Gennaretti et al., 2014
Payette, 2007z
D'Arrigo et al., 2003, 2013
Icefields
Idaho
Coppermine
Thelon
Quebec
Quebec
Northern Quebec
Labrador
68.39
59 to 62
65 to 70
49.02 to
50.59
ICE
52.16
IDA
40 to 45
COP
67.14
THE
64.02
QUEx
57.30
QUEw 57.30
NQU
55 to 60
LABrec 56.33 to
57.58
Eurasia
Scotland
Pyrenees
W Alps - Lotschental
E Alps - Tyrol
Jaemtland
€s
Tjeggelvas,Arjeplog, & Ammarna
composite
North Fennoscandia
57.08
42 to 43
46.5
47.30
63.30
65.54 to
66.36
EFmean 66 to 69
3.44
0 to 1
9
12.30
13.25
16.06 to 18.12
1200e2010
1260e2005
755e2004
1053e2003
783e2011
1200e2010
TRW/BI
MXD
MXD
MXD
MXD
MXD
STD/RCS
RCS
RCS
RCS
RCS
RCS
Rydval et al., 2017
~a
n et al., 2012
Dorado-Lin
Büntgen et al., 2006
Schneider et al., 2015
Zhang et al., 2016
Linderholm et al., 2015
19 to 32
750e2010
MXD
RCS
Forfjorddalen
Tatra
Mt Olympus, Greece
South Finland
Khibiny (Kola)
FORF
TAT
MOG
SFIN
KOL
15.43
19 to 20
22.37
28.19
33.13 to 34.15
978e2005
1040e2010
1521e2010
760e2000
821e2005
MXD
TRW
MXD
MXD
RW/BI
RCS
RCS
RCS
RCS/STD
Esper et al., 2014, Matskovsky and Helama,
2014
McCarroll et al., 2013
Büntgen et al., 2013
Klesse et al., 2015
Helama et al., 2014
McCarroll et al., 2013
Polar Urals
Yamal
Asia Grid 1
POLx
YAM
Grid1
65.40
69.54
60.15 to 68.15
891e2006
750e2005
817e1989
MXD
TRW
mix
RCS
RCS-SF
RCS/STD
Schneider et al., 2015
Briffa et al., 2013
Cook et al., 2013
Asia Grid 2
Grid2
70.15 to 78.15
827e1989
mix
RCS/STD
Cook et al., 2013
Asia Grid 10
Grid10
60.15 to 68.15
937e1989
mix
RCS/STD
Cook et al., 2013
Asia Grid 11
Grid11
70.15 to 78.15
937e1989
mix
RCS/STD
Cook et al., 2013
Kyrgyzstan
KYR
75.09 to 78.11
1689e1995
TRW/MXD
STD
Wilson et al., 2007z
Mangazeja
Asia Grid 3
MAN
Grid3
82.18
80.15 to 88.15
1328e1990
800e1989
MXD
mix
RCS
RCS/STD
Schneider et al., 2015
Cook et al., 2013
Asia Grid 12
Grid12
80.15 to 88.15
800e1989
mix
RCS/STD
Cook et al., 2013
Altai MXD
Asia Grid 4
ALT
Grid4
88.00
90.15 to 98.15
750e2007
800e1989
MXD
mix
RCS
RCS/STD
Schneider et al., 2015
Cook et al., 2013
Asia Grid 13
Grid13
90.15 to 98.15
1024e1989
mix
RCS/STD
Cook et al., 2013
Mongolia
Taymir
Asia Grid 5
OZN
TAY
Grid5
99.04
931e2005
102.00
755e1997
100.15 to 108.15 800e1989
TRW
TRW
mix
RCS
RCS
RCS/STD
Davi et al., 2015
Jacoby et al., 2000y
Cook et al., 2013
Asia Grid 14
Grid14
100.15 to 108.15 1396e1989
mix
RCS/STD
Cook et al., 2013
Asia Grid 6
Grid6
110.15 to 118.15 800e1989
mix
RCS/STD
Cook et al., 2013
Asia Grid 15
Grid15
110.15 to 118.15 1396e1989
mix
RCS/STD
Cook et al., 2013
Asia Grid 7
Grid7
120.15 to 128.15 1024e1989
mix
RCS/STD
Cook et al., 2013
SCOT
PYR
ALPS
TYR
JAEM
TAA
68.47
48 to 49
40.09
62.19
67.38 to
67.50
66.51
67.32
40.15 to
46.15
40.15 to
46.15
48.15 to
54.15
48.15 to
54.15
41.36 to
42.11
66.42
40.15 to
46.15
48.15 to
54.15
50.00
40.15 to
46.15
48.15 to
54.15
51.15
72.01
40.15 to
46.15
48.15 to
54.15
40.15 to
46.15
48.15 to
54.15
40.15 to
46.15
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
5
Table 1 (continued )
Site Name
Code
Asia Grid 16
Grid16 48.15 to
54.15
Grid8
40.15 to
46.15
Grid17 48.15 to
54.15
Grid9
40.15 to
46.15
Grid18 48.15 to
54.15
NJAP
43 to 51
YAK
67.27 to
70.33
Asia Grid 8
Asia Grid 17
Asia Grid 9
Asia Grid 18
North Japan
Yakutia
Latitude
Longitude
Time Span
Proxy Type Detrending Citation
120.15 to 128.15
1510e1989
mix
RCS/STD
Cook et al., 2013
130.15 to 138.15
1510e1989
mix
RCS/STD
Cook et al., 2013
130.15 to 138.15
1510e1989
mix
RCS/STD
Cook et al., 2013
140.15 to 148.15
1510e1989
mix
RCS/STD
Cook et al., 2013
140.15 to 148.15
1510e1989
mix
RCS/STD
Cook et al., 2013
142 to 145
142.37 to 150.17
1640e1993
1342e1994
TRW/MXD
TRW
STD
RCS
D'Arrigo et al., 2015
Hughes et al., 1999y
remain at 750 CE, the limits of our reconstruction. While all the
tree-ring chronologies and reconstructions have significant and
substantial correlations with local temperatures in one or more
months, locations that include MXD and BI data overall have higher
correlations compared to sites composed of tree-ring width data
alone (Fig. 2; see also Wilson et al. (2016)).
Fig. 1. Spatial and temporal distribution of sites by proxy type. Top panel (A) shows all
the records available during the best replicated (most recent, 1750 to 1988 CE) nest,
while panel (B) shows the distribution of available proxy sites at 1000 CE. Sites with
tree-ring width (TRW) data only are shown in green, while sites that have MXD, BI, or a
mix of proxy types are shown in blue. (C) Shows the total number of series through
time.
2.2. Observational data
Our target field (predictand) for our temperature reconstruction
is the interpolated hybrid (surface and satellite information)
version of HadCRUT4 from Cowtan and Way (2014). The original
HadCRUT4 (Morice et al., 2012) consists of monthly temperature
anomalies relative to the mean of the 1961 to 1990 CE period on a
regular 5 latitude/longitude grid and combines CRUTEM4 (Jones
et al., 2012) over land with HADSST3 (Kennedy et al., 2011a, b)
for the oceans. Use of the Cowtan and Way (2014) dataset provides
several advantages . First, this dataset seeks to compensate for
observational coverage bias and provides gridded estimates of
monthly temperature at high latitudes, including portions of our
target reconstruction region north of 40 N. Second, it is spatially
and temporally complete, allowing us to use the same calibration
and validation periods for our reconstruction at every location in
the field, which in turn permits straightforward comparisons of
reconstruction skill. Following Wilson et al. (2016), we use May
through August (MJJA) mean temperature anomalies as our target
variable, as this season provides a network-wide balance across the
diverse site-local monthly or seasonal climate responses of the
individual predictor series (Wilson et al., 2016).
Fig. 2. Kernel probability density estimate (Bowman and Azzalini, 1997) for the
highest local seasonal or monthly coefficient correlations as a function of proxy type.
Data are from Wilson et al. (2016)). The density estimate is calculated for a support of
[1,1] and the values contributing to the distribution are indicated by symbols along
the x-axis.
6
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
2.3. Reconstruction and statistical methodology
As a prelude to our climate field reconstruction, we assess the
spatial characteristics of the temperature signal across our
network. We first calculate the Pearson Product Moment correlation between each series and the local gridded MJJA temperatures
from Cowtan and Way (2014) as a measure of the correspondence
between tree growth and local gridded temperatures. We also
calculate, for each site, the field correlation between the individual
predictor series and the entire MJJA and annual mean temperature
field, using both the original data as well as first-differenced series.
For assessing the association between the target field and the
predictor network, we follow Schneider et al. (2015) and compute
the median correlation coefficient between the temperature record
at each grid box and all the predictor time series within 2000 km of
the centroid (see below; Briffa and Jones, 1993; Jones et al., 1997;
Cook et al., 2013). We perform this procedure for both the best
replicated part of our predictor network (1750e1988 CE) and at
1000 CE in the midst of the Medieval epoch. Collectively, these
statistical assessments provide an evaluation of the climate signal
embedded in the predictor network through time and space.
We use Point-by-Point Regression (PPR; Cook et al., 1999) to
reconstruct the MJJA surface temperature anomaly field north of
40 N using our predictor network of tree-ring proxy chronologies
and temperature reconstructions. We follow the method as
developed, tested, and described by Cook et al. (1999, 2010a, 2013).
PPR incorporates the spatial structure of the predictor network and
predictand field and confines the potential region of influence of
the predictors to a distance estimated from the underlying correlation structure of the temperature field. PPR proceeds by calculating a nested multivariate regression model for each grid point in
the target field with the predictors restricted to those within some
radius from the grid point centroid. We adopted a dynamic search
radius for each grid point in the target field, first identifying predictor series within 1000 km. If no chronologies were found within
1000 km, the radius was allowed to expand in 500 km increments
up to a maximum of 2000 km to find predictors. These distances are
based on the decorrelation decay as a function of distance in the
target field data (Cowtan and Way, 2014) and are also consistent
with the findings of other studies (Briffa and Jones, 1993; Jones
et al., 1997; Cook et al., 2013). If no predictors were found within
2000 km, then no climate reconstruction was produced for that
grid point. In evaluating our methods, we found that this dynamic
search radius provides an optimal balance between maximizing the
number of grid points available for reconstruction while allowing
the local predictor series in data-dense regions (for instance, Fennoscandia and western Europe) to provide the paleoclimate information for their neighborhood of grid points.
A multivariate regression model was calibrated for each grid
point and its associated predictors over the period 1945 to 1988 CE
(the latest date for which all chronologies had data) and then
validated on withheld observational data over the period 1901 to
1944 CE (Michaelson, 1987). We also checked the sensitivity of our
reconstruction to this choice of calibration and validation periods
by swapping them and assessing cross-validation. As the number of
predictor series declines back through time, the model is newly
calibrated and validated at each change in sample depth. In our
reconstruction, we use the individual series themselves as predictors as opposed to the leading principal components (PCs). In
our sensitivity tests of the PPR method, we discovered that using
PCs from a relatively sparse network of chronologies and reconstructions, combined with the expansive target field, created
clearly artificial inhomogeneities or discontinuities when predictor
numbers declined. Using the individual predictor time series
themselves in a stepwise regression model with an adjusted R2
entry rule (Meko, 1997) ameliorated such discontinuities. Model
skill was assessed using the calibration R2c (adjusted for the number
of predictors), the validation R2v , the Reduction of Error (RE) and the
Coefficient of Efficiency (CE) (c.f. Cook et al., 1999; Wilson et al.,
2006). In addition to the annual maps of reconstructed temperatures, we calculate an extratropical Northern Hemisphere mean
MJJA time series using a latitude-weighted average of all the
reconstructed grid cells where reconstructed values are available
back to at least 1000 CE and RE is greater than zero.
Following Masson-Delmotte et al. (2013), we calculated the
difference between reconstructed Medieval Climate Anomaly
(MCA) and Little Ice Age (LIA) temperatures by taking the difference
between the mean values over the field for 950 to 1250 CE (MCA)
and 1450 to 1850 CE (LIA). As there is no single accepted definition
of these two periods (Hughes and Diaz, 1994; Bradley et al., 2001;
Matthews and Briffa, 2005; Seager et al., 2008; Mann et al.,
2009), we also tested the sensitivity of the calculated MCA-LIA to
Fig. 3. Local correlations between tree-ring proxy chronologies and local temperature data. (A) Correlations reported by the original authors for local correlations with the optimal
seasonal or monthly window (see Wilson et al., 2016; their Table 1). (B) Correlations between each proxy series and the local May through August (MJJA) temperature data from
Cowtan and Way (2014) used here.
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
differences in the time definition of these periods. We estimated
the temperature response to tropical explosive volcanism (Robock,
2000; Ammann and Naveau, 2003) by calculating the composite
mean anomaly using superposed epoch analysis (e.g. Haurwitz and
Brier, 1981). Event years were extracted from updated estimates of
Common Era volcanic forcing from Sigl et al. (2015), here selecting
those years corresponding to an estimated maximum negative
event forcing (in W m2) with a magnitude at least as large as
Krakatoa in 1883.
3. Results
3.1. Network climate signal
All our predictor series show significant and typically high
correlations with local summer temperatures over one or several
months (Figs. 2 and 3; Wilson et al. (2016)). At sites where the
7
highest local summer temperature signal in the series is confined to
one or two months e for example, at Yakutia in Russia e local
correlations with the broader MJJA season are lower (r ¼ 0.52, p <
0.05 for July vs. r ¼ 0.09, p > 0.05 for MJJA at Yakutia). The highest
individual monthly/seasonal site local temperature correlations
(Wilson et al., 2016) range from r ¼ 0.39 to r ¼ 0.84 (mean r ¼ 0.63),
while correlations with local MJJA temperatures range from r ¼
0.04 to r ¼ 0.78 (mean r ¼ 0.43). The highest correlations with local
MJJA temperatures are in Fennoscandia and north central Russia,
with strong local temperature signals also evident in Scotland, the
Alps, the Pyrenees, the Altai, and Japan. In North America, MXD
chronologies from western Canada and the northern treeline have
the strongest MJJA signals. Tree-ring width only chronologies in
North America have a generally weaker association with their local
MJJA instrumental temperature than MXD or mixed proxy sites
(D'Arrigo et al., 1992; Jacoby and D'Arrigo, 1995; D'Arrigo et al.,
2009; Andreu-Hayles et al., 2011; Anchukaitis et al., 2013).
Fig. 4. Field correlations between each tree-ring site (indicated by the black stars) and the MJJA mean temperature field from Cowtan and Way (2014). Labels correspond with the
site codes from Table 1. Around each site the black range ring indicates a radius of 2000 km. Only Pearson Product Moment correlation coefficients (r) significant at p < 0.05 as
adjusted for autocorrelation (Trenberth, 1984) are plotted.
8
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Field correlations between the predictor series and the full
MJJA temperature field (Fig. 4) suggest that the chronologies and
temperature reconstructions reflect climate variability over
many hundreds or thousands of kilometers, with some exceptions at those sites where the local proxy response to MJJA is
already weak (e.g. locations in east Asia, Yakutia, and ring
width-only chronologies from the North American treeline). The
spatial extent of the large-scale correlation structure is partially
related to the common positive trends in predictors and temperatures during the 20th century, as temporarily removing
these trends by first differencing both the field and the predictors (Fig. 5) reduces the regions of positive correlations to
between 500 and 2000 km. In some case e e.g. the Idaho (USA)
chronology e a significant interannual correlation with the MJJA
temperature field is entirely absent, indicating the summer association there is driven by common trends and a narrow local
monthly temperature response. Annual temperatures are often a
target for temperature reconstruction (e.g. Esper et al., 2002;
D'Arrigo et al., 2006; Mann et al., 2008, 2009); however, correlations with the annual temperature field herein are uniformly
lower and many chronologies show no significant association
with annual mean temperatures (Fig. 6). In general, higher
correlations with annual mean temperatures are observed for
those sites with higher local correlations with MJJA temperatures (rMJJA;annual ¼ 0:73, p < 0.01). The association between the
annual signal and the best (highest) monthly or seasonal local
climate correlations is weaker (rbest;annual ¼ 0:44, p < 0.01),
indicating it is not a strong climate signal alone that corresponds
Fig. 5. Field correlations between each tree-ring site (indicated by the black stars) and the MJJA mean temperature field from Cowtan and Way (2014) after first-differencing each
variable to reduce the influence of common trends. Labels correspond with the site codes from Table 1. Around each site the black range ring indicates a radius of 2000 km. Only
Pearson Product Moment correlation coefficients (r) significant at p < 0.05 as adjusted for autocorrelation (Trenberth, 1984) are plotted.
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
with a useful annual proxy, but rather a strong and broad seasonal climate response.
Fig. 7 shows the statistical relationships between the predictand
target field and predictor network. Grid points in Asia, Fennoscandia, and Northern Europe have 15 or more proxy sites that can
serve as predictors over the best replicated epoch (1750e1988 CE).
In northwestern North America as many as 10 predictors are within
2000 km of the grid centroids. In contrast, grid points in southern
Europe and eastern North America have many fewer predictors
available, in some cases only a single series. For the best replicated
portion of the reconstruction, 1750 to 1988 CE, the median correlation between observational MJJA temperatures at each grid point
and the predictors within 2000 km of that grid point range from
r ¼ 0.19 in northeastern Russia to r ¼ 0.66 in Fennoscandia. In
9
general, the median grid correlations are highest where they are colocated with one or more chronologies (e.g. interior British
Columbia) and regions with clusters of strong MJJA temperature
proxies (e.g. Fennoscandia, the Alps and Pyrenees, northern North
American treeline). By 1000 CE, when the number of predictors is
reduced to 23, with only 3 of these in North America, the possible
reconstruction domain is reduced as substantially fewer predictors
are available for each grid point reconstruction. Nevertheless, grid
median correlations between observed temperatures and the
available predictors remain significant and of similar magnitude to
the better replicated recent portion of the reconstruction. This is
because predictors with strong temperature signals remain in the
Alps, Icefields in British Columbia, the Gulf of Alaska, Fennoscandia,
northern Russia, and the Altai. The reconstruction at 1000 CE and
Fig. 6. Field correlations between each tree-ring site (indicated by the black stars) and the annual mean temperature field from Cowtan and Way (2014) after first-differencing each
variable to reduce the influence of common trends. Labels correspond with the site codes from Table 1. Around each site the black range ring indicates a radius of 2000 km. Only
Pearson Product Moment correlation coefficients (r) significant at p < 0.05 as adjusted for autocorrelation (Trenberth, 1984) are plotted.
10
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
earlier therefore relies on a reduced number of predictors, but those
that remain contain a significant and substantial temperature signal.
3.2. Field reconstruction, calibration, and validation
The full NTREND spatial reconstruction consists of 1239 yearly
fields of MJJA temperature anomalies covering up to a potential 792
grid points each year (40e90 N, 180 to 180 E). Fig. 8 shows the
length of the reconstructed temperature series at each grid point in
our domain. In practice the number of grid points with a value in
each year of the reconstruction is less than that potential
maximum, as the full network can only support reconstruction at
85% (n ¼ 701) of the total grid cells in the domain, declining to 51%
(n ¼ 401) by 1000 CE and 24% (n ¼ 190) by 750 CE. The shortest
reconstructions are for ocean grid points in the northwestern
Atlantic and northeastern Pacific, where in both cases a single and
relatively short chronology is the only source of proxy information.
Other relatively short portions of the field reconstruction include
the central northern treeline in Canada, Japan, northeastern Russia,
and the southernmost part of the Eurasian domain in the Mediterranean. By contrast, along the Eurasian treeline in central
northern Russia, throughout the Nordic region, and in western
Europe, it is possible to reconstruct more than a millennium of past
temperatures.
Fig. 9 shows reconstruction skill (the adjusted R2c , R2v , RE, and CE)
for the best replicated (n ¼ 54) period of the predictor network,
1750 to 1988 CE. Significant skill is observed over the entire
domain, but is clearly highest closest to those predictors with the
highest correlations to MJJA temperatures, especially where MXD
or BI data are available. Grid cells more distal from the predictor
network, including cells over the oceans, or where only a single or
relatively weak predictor is available, show lower explained variance and in some cases lack positive RE and CE scores. Validation R2v
scores are lower but largely mirror the calibration R2c , with the
exception of the eastern Mediterranean and Black Sea region, east
Asia, and western parts of Central Asia. A similar phenomenon of
lower R2v was observed by Cook et al. (2013) for Asia, due at least in
part to the lack of instrumental climate data from these regions
during the reconstruction model validation period (Jones et al.,
2012; Harris et al., 2013; Cook et al., 2013; Cowtan and Way,
2014). Lack of instrumental data likely confounds out-of-sample
validation in the eastern Mediterranean prior to the 1930s (c.f.
Touchan et al., 2014). Skillful regions with RE and CE scores greater
than 0 are more spatially confined but likewise show skill with
respect to these metrics in regions where chronologies are present
or abundant, with the exception once again of the regions
mentioned previously. R2c values range from ~0.0 to 0.78, R2v values
range from ~0.0 to 0.76, RE up to 0.72, and CE up to 0.70. A cross-
Fig. 7. Proxy and target grid characteristics. Panels (A) and (C) show the number of tree-ring sites within 2000 km for each grid point in the target field (Cowtan and Way, 2014),
during the best replicated (modern) nest (A) and at 1000 CE (C), respectively. Black circles indicate the location of the available tree-ring sites during in each time period. Panels (B)
and (D) show the median value at each target field grid point for the Pearson Product Moment correlation between the MJJA temperatures at that grid point and all the tree-ring
chronologies within 2000 km of that grid, during the best replicated (modern) nest (B) and at 1000 CE (D), respectively (see Schneider et al., 2015).
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Fig. 8. Reconstruction length. The length (in years) of the reconstruction at each target
grid point.
validation (not shown) interchanging the time periods used for
calibration and validation reveals that the reconstruction's skill
characteristics are largely insensitive to the choice of these periods.
11
By 1000 CE, the reduction in the number of predictors and a
contraction in their spatial distribution influences both the number
of grid points reconstructed and the spatial patterns of skill
(Fig. 10). The loss of the northern treeline MXD chronologies in
North America reduces the reconstructed regions of the continent
in the west to coastal Alaska, the Pacific Northwest, Interior British
Columbia and parts of Alberta, and in the east to Quebec,
Newfoundland, and Labrador. Likewise, the loss of the Scottish and
Pyrenees chronologies no longer allows for reconstruction of
temperatures over the British Isles and the Iberian peninsula. The
lack of Japanese, coastal Russia, and east Asia series at 1000 CE leads
to a contraction of the reconstructed spatial domain in the east.
However, skillful temperature reconstructions persist over parts of
northwestern North America, Fennoscandia, northern Russia, the
Alps, and the Altai. R2c values range from ~0.0 to 0.73, R2v values
range from ~0.0 to 0.70, RE up to 0.65, and CE up to 0.61. Interchanging the calibration and validation periods has a minor influence on field skill at 1000 CE, with n ¼ 229 grid points with RE > 0
RE > 0 for a late calibration (1945e1988 CE), and n ¼ 213 for an early
calibration (1901e1944 CE).
Fig. 11 shows domain-wide reconstruction and skill metrics over
time. For the time span of the full predictor network (1750e1988
CE), we can reconstruct a temperature anomaly value for 88.5% of
the grid points (n ¼ 701) in our extratropical Northern Hemisphere
domain. Of those 701 grid points, 63% have RE > 0, and 37% have
CE > 0. These skill percentages remain remarkably stable (RE, 56%e
65%; CE, 34%e37%) even as the number of reconstructed grid points
with a reconstructed value declines back through time, a consequence of the shrinking predictor network. Only when the number
of reconstructed grid points declines precipitously in the earliest
10th century, falling to 38% of the target domain in 905 CE, do the
percent of grid points with RE and CE greater than zero increase
Fig. 9. Reconstruction skill for the best replicated (modern) nest. Panels show spatial patterns of skill metrics for the best replicated nest (1750e1988 CE), as evaluated for the
adjusted calibration R2 , the validation R2 , the reduction of error (RE), and the coefficient of efficiency (CE). Available tree-ring sites during this nest are indicated by black circles.
12
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Fig. 10. Reconstruction skill at 1000 CE. Panels show spatial patterns of skill metrics for the reconstructed field at 1000 CE in the midst of the Medieval epoch, as evaluated for the
adjusted calibration R2 , the validation R2 , the reduction of error (RE), and the coefficient of efficiency (CE). Available tree-ring sites at 1000 CE are indicated by black circles.
substantially. These patterns indicate that the reduction in the
number of reconstructed grid cells comes at the cost of locations
with already marginal skill scores, while the core reconstruction
regions associated with sensitive predictor series persist through
much of the length of the reconstruction back to 750 CE.
Fig. 11. Aggregate temporal reconstruction skill. The percent of all target grid points
that are able to be reconstructed for a given year is shown by the black line. For those
grid points with a reconstructed value in a given year, the red and blue lines show the
percent of those grid points with RE and CE greater than zero.
3.3. Large-scale mean temperature anomalies and climate forcing
We computed a mean summer temperature anomaly series
from our domain by calculating a latitude-weighted mean of all
gridded values where a reconstruction is available back to 1000 CE
and RE is greater than zero (Fig. 12). This time series is highly and
significantly correlated with the comparable observational temperature anomalies averaged over those same grid points
(1850e1988, n ¼ 139, r ¼ 0.78 , p ≪ 0.001). The mean series indicates a broad warm period from at least 750 CE to the early 1400s,
with maximum values centered around the late 900s, the late
1000s and 1100s, and the individual warmest years of the Medieval
epoch in 790, 990, 995, 1014, 1016, and 1168 CE. 1168 CE is the
warmest year in our reconstruction (750e1988 CE), although
matched by values in the middle of the 20th century and then
exceeded during the comparable filtered instrumental record in the
early 21st century. Temperatures decline in the late 13th century,
coincident with a series of tropical volcanic eruptions (Crowley,
2000; Gao et al., 2008; Schmidt et al., 2012; Sigl et al., 2015) and
the Wolf solar irradiance minimum, before warming again during
the 14th century. Temperatures then decline sharply in the early
1400s, slightly before the significant volcanic eruptions in the
€rer solar irra1450s (Gao et al., 2006; Sigl et al., 2015) and the Spo
diance minimum (1460e1550 CE). The mid-1400s through the
mid-1800s show cooler conditions during the LIA (Matthews and
Briffa, 2005; Masson-Delmotte et al., 2013), with local minima in
the 1450s, the late 1500s and earliest 1600s, the late 1600s, and the
early 1800s. Many of the field-mean coldest years in the reconstruction, including 1259, 1453, 1601, 1643, 1783, 1810, 1817, and
1836 CE are associated with or follow closely after major tropical or
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Fig. 12. Filtered latitude-weighted mean hemisphere MJJA temperature anomaly
reconstruction and target MJJA observational time series. Spatial mean values for both
the reconstruction (black) and target (red) MJJA fields are calculated from the set of all
grid points that have reconstructed values back to at least 1000 CE and which have an
RE score greater than zero at 1000 CE (n ¼ 229n ¼ 229) and are weighted by latitude.
Uncertainty in the reconstruction is indicated by the grey shading, and is calculated as
the mean latitude-weighted local mean squared error of validation. The reconstruction
and target MJJA temperature series are significantly correlated over their common
interval (1850e1988, n ¼ 139 n ¼ 139, r ¼ 0.78 r ¼ 0:78, p ≪ 0.001).
Northern Hemisphere volcanic eruptions (Briffa et al., 1998a; Sigl
et al., 2015). In several instances extremely cold years in our
13
reconstruction occur or persist at least one or two years after
eruptions (e.g. 1643 and 1644, 1816 and 1817, 1836 and 1837 CE),
consistent with the findings of Esper et al. (2013b). The coldest year
in our reconstruction is 1601 CE, which also agrees with a prior
temperature reconstruction based solely on MXD data by Briffa
et al. (1998a), and which follows the eruption of Huaynaputina in
Peru in 1600 CE (Verosub and Lippman, 2008). Our ten coldest
years include at least 2 (1699 and 1867 CE) that do not appear to be
associated with a known volcanic eruption (Briffa et al., 1998a).
Our filtered time series is highly and significantly correlated
(750e1988 CE, n ¼ 1239, r ¼ 0.71, p ≪ 0.01) with the index
reconstruction from Wilson et al. (2016) and perhaps unsurprisingly has many of the same features - a broad warm period during
the Medieval period until the early 1400s, the LIA from the 15th
through early 19th century, warming out of the LIA, a warm mid20th century, cool 1970s, and recent warming (Fig. 13a). There are
epochs where our mean time series and that of Wilson et al. (2016)
are less strongly correlated (Fig. 13b); interestingly, the rapid
returns to higher correlation values appear to be associated with
the timing of major individual or clusters of volcanic events, suggesting that strong radiative forcing due to volcanism may impose a
common spatial forcing, causing the Wilson et al. (2016) mean index reconstruction and the spatial mean from our climate field
reconstruction to converge.
Our Northern Hemisphere extratropical time series shows associations with large-scale radiative forcing changes during the last
millennium (Fig. 14; Schmidt et al., 2012). Colder periods in the late
13th and early 14th, mid-15th, and 19th century occur at the same
time as large explosive volcanic eruptions and solar minima (Mann
et al., 1998; Crowley, 2000; Shindell et al., 2001a, 2003; Wagner and
Fig. 13. Comparison between time series reconstruction from (blue; Wilson et al., 2016) and the filtered weighted global mean MJJA temperature reconstructed here (black). Time
series are shown in (A), and a running correlation (50 year window, 1 year increment) is plotted in (B). The full overall correlation between the two series (750e1988 CE, r ¼ 0.71) is
indicated by the dashed red line in (B).
14
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Fig. 14. Radiative forcing and reconstructed Northern Hemisphere warm-season
temperatures from this study during the last millennium. All forcing series are those
compiled by Schmidt et al. (2012) for PMIP3 simulations of the last millennium
(version 1.1). (A) Volcanic forcing following Gao et al. (2008) (black/grey, GRA) and
Crowley et al. (2008) (blue/light blue, CEA), with individual years as lighter lines and
30-year Gaussian smoothed values in heavy lines. Note that the magnitude of some
individual events exceeds the y-axis limits. (B) Northern Hemisphere mean MJJA
temperature anomaly time series as described in the text (black line) and corresponding observed temperatures for the same grid points (red line). Here both
reconstructed and observed values have been smoothed with a 30 year Gaussian filter.
(C) Solar forcing relative to the period 1976 to 2006 CE, with the pink shaded region
showing the range of the forcing reconstructions compiled by Schmidt et al. (2012)
including Delaygue and Bard (2011), Muscheler et al. (2007), Steinhilber et al. (2009)
and Vieira and Solanki (2010). Major solar minima are labeled. (D) Forcing due to
land use change from Kaplan et al. (2011) (KK10) and Pongratz et al. (2008) (PEA). (E)
Well-mixed greenhouse gas forcing.
Zorita, 2005; Ammann et al., 2007; Breitenmoser et al., 2012;
Anchukaitis et al., 2013; PAGES2k, 2013). Warming after the middle of the 19th century is consistent with a reduced number of
volcanic eruptions, increasing insolation, and the rapid rise in
greenhouse gases (Andronova and Schlesinger, 2000; Zwiers and
Weaver, 2000; Gillett et al., 2012; Jones et al., 2013; Estrada et al.,
2013). Although the timing of several epochs of colder temperatures appear to align with solar minima e for instance, the late
13th/early 14th century and the Wolf Minimum or the mid-15th
€rer Minimum e the correlations between our hemicentury Spo
sphere mean reconstruction and estimates of past solar variability
(Schmidt et al., 2012) are low (range, r ¼ 0.10 to r ¼ 0.29, p < 0.01).
At centennial timescales, however, there is evidence that solar
variability may play a more substantial role. Wavelet coherence
(Fig. 15; Grinsted et al., 2004) between our hemisphere mean
reconstruction and last millennium total solar irradiance estimates
assembled by Schmidt et al. (2012) shows high and stable coherence and consistent phasing at bi-centennial time scales (194e222
year periods), which bracket and include the ~206 year ‘de Vries’
(or Suess) solar cycle (Stuiver and Braziunas, 1993; Wagner et al.,
2001) and are likely related to the reconstructed temperature
response to the major solar minima. The spectral signal of the
bicentennial de Vries cycle has been recognized in numerous treering chronologies and temperature reconstructions (c.f. Raspopov
et al., 2008; Breitenmoser et al., 2012; Ogurtsov et al., 2016), and
Emile-Geay et al. (2013) also identified bi-centennial periodicity in
a reconstruction of eastern tropical Pacific sea surface temperatures. Phase relationships between our hemisphere mean temperature and the total solar irradiance time series suggest a decadalscale lag of ~11 years (range, 5e20 years), with solar changes
leading temperature anomalies, consistent with both climate
modeling and prior analysis of tree-ring chronologies and solar
variability (Rind et al., 1999; Waple et al., 2002; Breitenmoser et al.,
2012). Both the hemisphere mean as well as the spatial grid point
sensitivity to solar variability (C/W m2) are extremely uncertain in
this analysis, however, as this quantity is highly sensitive to the
choice of solar reconstruction (Schmidt et al., 2012).
Medieval Climate Anomaly (MCA; 950 to 1250 CE) temperatures
compared against those during the Little Ice Age (LIA; 1450 to 1850
CE) show warmer temperature during the MCA at ~90% of the grid
points with minimally skillful (RE > 0) reconstructed values available back to 950 CE (Fig. 16). Colder Medieval temperatures are
reconstructed over parts of the Altai and Central Asia, and are
associated with tree-ring width chronologies from Mongolia that
show reduced growth in the 900s and 1100s, despite warmer
conditions in the 11th century (Davi et al., 2015) and an MXD
chronology from the Altai displaying a cold Medieval epoch and
Fig. 15. Wavelet coherence (Torrence and Compo, 1998; Grinsted et al., 2004) between our Northern Hemisphere mean MJJA temperature anomaly time series and solar forcing
variability from Vieira and Solanki (2010). Arrows indicate the phase of the relationship and for clarity are plotted only where coherence exceeds 0.65. In-phase signals point directly
to the right of the plot. Values above the cone of influence (COI; black curve) are potentially influenced by edge effects at that time period and scale.
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
Fig. 16. Medieval Climate Anomaly (MCA; 950e1250 CE) vs Little Ice Age (LIA;
1450e1850 CE) mean temperature anomaly fields (MCA-LIA). Only grid points with
values reconstructed at RE > 0 at 1000 CE are shown.
warm LIA (Schneider et al., 2015). Other grid cells that show a
colder Medieval period tend to be distal from the predictor network
e for instance in central Greenland e and must be treated with
caution. Defining a different MCA or LIA in this case has relatively
little effect on the percentage of grids showing warmer vs. colder
conditions, as the cooler Central Asia grid cells and marginal cells in
Greenland, central Canada, the southern Caspian Sea, and the
northeastern Pacific remain irrespective of the specific date range
applied. The precise boundaries for both MCA and LIA are, in any
case, both arbitrary and uncertain (Hughes and Diaz, 1994; Bradley
et al., 2001; Matthews and Briffa, 2005; Seager et al., 2008). The
cause of apparently extremely high (~3 C) Medieval temperatures
in several grid points in northeastern North America is discussed
below.
Composite mean MJJA temperature anomaly fields following
major volcanic eruptions show coherent, broad-scale cooling
associated with large tropical eruptions (Fig. 17). 96% of grid points
show composite mean colder temperatures compared to the three
years prior to the 20 large eruptions considered here. Similar to the
MCA-LIA difference discussed above, regions that apparently have
an overall composite warming response to volcanic eruptions are
largely on the margins of the reconstruction domain, away from the
predictor grid, and over the ocean, central Greenland, and the
southern Caspian Sea. The coldest grid point composite mean
is 1.61 C, and the mean composite response across all grid points
and all eruptions is 0.44 C. Closer examination of individual
eruption events (not shown) finds that for some regions, postvolcanic cooling may persist for several years and maximum cold
anomalies may be 1 or 2 years after the year of the eruption itself,
consistent with observations of other regional temperature reconstructions (D'Arrigo et al., 2013; Cook et al., 2013; Esper et al.,
2013b; Davi et al., 2015; Linderholm et al., 2015; Schneider et al.,
2015; Wilson et al., 2016). If we consider large Northern Hemisphere high-latitude eruptions only (Fig. 18), the large-scale
response is likewise toward cold anomalies overall: 89% of grid
15
Fig. 17. Composite mean reconstructed temperature anomalies following major tropical volcanic eruptions (from Sigl et al., 2015). Eruption years in the composite (n ¼ 20)
are those with a global forcing magnitude equal to or larger than that associated with
Krakatoa (1884), and include 916, 1108, 1171, 1191, 1230, 1258, 1276, 1286, 1345, 1453,
1458, 1595, 1601, 1641, 1695, 1809, 1815, 1832, 1836, and 1884 CE. Event anomalies are
calculated by first subtracting the global field mean over the 3 years prior to the
eruption. Only grid points with RE > 0 in an event year are averaged to form the
composite and only those grid points with values for at least 6 eruptions are plotted.
points have a composite anomaly less than zero. Over the entire
field the mean composite response is 0.39 C and the maximum
cold composite anomaly is 2.31 C. There is also spatial structure
to the temperature anomalies, with the coldest composite conditions over Alaska and the Bering Strait, northeastern North America, parts of western Europe, and central northern Russia,
suggestive of a dynamical, in addition to direct radiative, influence
Fig. 18. Composite mean reconstructed temperature anomalies following major
Northern Hemisphere high latitude volcanic eruptions (from Sigl et al., 2015). Eruption
years in the composite (n ¼ 5n ¼ 5) are those Northern Hemisphere eruption with a
global forcing equal to or larger than the magnitude associated with Katmai (1912),
and include 939, 1182 1210, 1783, and 1912 CE. Event anomalies are calculated by first
subtracting the global field mean over the 3 years prior to the eruption. Only grid
points with RE > 0 in an event year are averaged to form the composite and only those
grid points with values for at least 2 eruptions are plotted.
16
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
of large-magnitude high latitude eruptions (Robock, 2000; Oman
et al., 2005; Stenchikov et al., 2006; Schneider et al., 2009;
Zanchettin et al., 2012; Pausata et al., 2015). However, the number of radiatively significant high latitude eruptions considered
here is smaller (n ¼ 5), and therefore this structure may appear due
to the limited sample size.
4. Discussion
4.1. Proxy data and predictor network
Our results here demonstrate that a relatively small (n ¼ 54)
network of proxy sites (Table 1, Fig. 1) with well-established
physically and ecologically reasonable climate signals (Figs. 3 and
7) can be used to reconstruct the large-scale summer temperature history of the extratropical Northern Hemisphere (Figs. 8e10,
12). While restricting the reconstruction to the higher latitudes of
one hemisphere and to only the growing season does not provide a
global annual estimate of temperature, it nonetheless accurately
reflects the geographic and biological signal that dominates the
predictors. Moreover, we have demonstrated here that the reconstruction preserves the signature and influence of external forcing
on the global energy balance. Skill in our reconstruction is, perhaps
not surprisingly, greatest in those locations where high quality
temperature-sensitive proxies are available (Figs. 9 and 10), and
declines at increasing distances from the predictors themselves. It
is clear we could realize substantial benefits in terms of increased
reconstruction skill and spatial extent by developing MXD and BI
chronologies from currently undersampled regions as well as
extending the length of existing MXD and BI chronologies.
Although large and useful multiproxy datasets have resulted from
community efforts by the paleoclimate community (PAGES2k,
2013), our analysis here demonstrates that a relatively small welldistributed network of highly sensitive millennium-length tree
ring chronologies provide skillful reconstructions over a large
extratropical region. Encouragingly, this means that rapid and
important gains could be made from the addition of a relatively
small number of new sites and the temporal extension and recollection of current sites known to contain a strong climate signal.
In particular, a greater number of long MXD and BI chronologies are
critically needed from North America (Figs. 1 and 10). Fulfilling this
need will require a collaborative and concerted effort to locate
subfossil materials and to measure density proxies, but the potential gain for Northern Hemisphere temperature reconstructions
of the Common Era would be substantial. MXD contains a stronger
temperature signal than TRW alone and is better able to accurately
resolve rapid temperature changes associated with volcanic eruptions (Fig. 2; Frank et al., 2007; D'Arrigo et al., 2013; Esper et al.,
2015). Continuing advances in blue intensity (BI) measurements
suggest some of the benefits of wood density analysis can be
realized without the expense and difficulty of analyzing MXD itself
(Campbell et al., 2007; Wilson et al., 2014; Rydval et al., 2014;
€ rklund et al., 2014, 2015) although the low-frequency characBjo
teristics of BI still require additional exploration.
Detrending and standardization issues for long chronologies
remain an ongoing challenge and persistent source of uncertainty
(e.g. Cook et al., 1995; Briffa and Melvin, 2011; Melvin and Briffa,
2008; Esper et al., 2012, 2016; Matskovsky and Helama, 2014;
Matskovsky and Helama, 2016). One surprising feature of the
epochal comparison between the MCA and LIA is the large (>3 C)
difference calculated for northeastern North America. This feature
is due to a single predictor, a black spruce (Picea mariana) tree-ring
width RCS chronology developed by Gennaretti et al. (2014). Other,
non-tree ring proxies from the region suggest a lower amplitude of
cooling between the MCA and the LIA, although issues related to
time uncertainty, transfer function calibration, and different seasonal climate signals complicate exact comparisons. A compilation
of Holocene paleoenvironmental data for the Arctic (Sundqvist
et al., 2014) suggests a range of values for MCA to LIA cooling in
northeastern North America and Greenland of 0e1.5 C, which is at
least a degree less than the magnitude inferred from the Quebec
MXD record. Alkenone SST reconstructions near Nova Scotia
(Keigwin et al., 2003) suggest a cooling of approximately 0.7 C. The
multiproxy PAGES Arctic2k reconstruction (McKay and Kaufman,
2014) has a whole-Arctic reconstructed MCA-LIA difference of
0.64 C for the same time periods used here. Finally, and perhaps
most importantly, a tree-ring oxygen isotope temperature reconstruction from the same site in Quebec (Naulier et al., 2015) shows a
substantially smaller estimated MCA-LIA difference of 0.4 C. It
seems likely that, despite the care taken in applying regional curve
standardization to the Quebec black spruce samples (Autin et al.,
2015) as well as its suitability with respect to other chronology
metrics (Esper et al., 2016), artifacts remain in this tree-ring width
chronology that unintentionally but artificially amplify the difference between MCA and LIA temperatures in this region.
More generally, detrending, the removal of non-climatic trends,
and therefore the retention of low frequency variability, remains an
important source of uncertainty in the amplitude of past temperatures reconstructed from tree rings (Cook, 1987; Briffa et al., 1996),
even when conservative detrending techniques have been applied.
While regional curve standardization and signal free methods have
been shown to be able to retain the full spectrum of low- and
medium-frequency variability, they are also subject to their own
uncertainties and assumptions (Melvin, 2004; Melvin and Briffa,
2008; Briffa and Melvin, 2011; Anchukaitis et al., 2013; Briffa
et al., 2013). It is in most cases not possible to know from calibration and validation statistics which detrending method yields the
true or most accurate low frequency signal (Cook, 1987; Cook and
kstis, 1990; Wilson et al., 2007). Possible approaches to this
Kairiu
problem include both ensemble and simulation-based methods
(e.g. Esper et al., 2007; D'Arrigo et al., 2011; Anchukaitis et al.,
2013), although these have not yet been applied to large and heterogeneous tree-ring proxy networks.
4.2. Radiative forcing and temperature history
Our reconstruction demonstrates coherent responses to radiative forcing in time and space (Figs. 14e18). Temporal features of
the reconstruction are associated with changes in solar irradiance,
large and/or clustered tropical volcanic eruptions, and the anthropogenic rise in well-mixed greenhouse gases. Temperatures decline
€rer and Maunder Minima, in
across the field during the Spo
particular, likely compounded in both cases by a series of volcanic
eruptions. Temperatures remained cold during the early 1600s, at
least in part due to the eruption of Huaynaputina in Peru in 1600 CE
(Verosub and Lippman, 2008). 1601 CE is the coldest year of our
entire reconstruction, as it was in the 600-year temperature
reconstruction by Briffa et al. (1998a) and the hemisphere mean
reconstruction by Wilson et al. (2016). 1601 CE was also one of the
coldest years in the Bayesian field reconstruction by Tingley and
Huybers (2013), as was 1453 CE, which is the 4th coldest year in
our study, associated with the eruption of Kuwae, Vanuatu (but see
Plummer et al., 2012; Cole-Dai et al., 2013). Interestingly, Tingley
and Huybers (2013) find 1642 CE amongst their coldest years,
whereas in our reconstruction it is 1643 CE that is exceptionally
cold (5th coldest in our reconstruction). In Wilson et al. (2016),
1641, 1642, and 1643 CE are all amongst the coldest 15 years of their
reconstruction. Tingley and Huybers (2013) also find that 1695 CE
was anomalously cold, whereas here it is indeed cold but unremarkable (0.60 C, 284th coldest). These differences highlight
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
extant uncertainties likely related to different reconstruction
methods, spatial skill and averaging, and the use of different
proxies (D'Arrigo et al., 2013; Esper et al., 2015), but also demonstrate that there is no evidence for a one-to-one correspondence
between inferred volcanic forcing from ice cores and the magnitude
of hemisphere-scale cooling. For instance, the eruption of Huaynaputina in 1600 is believed to have caused a smaller negative
radiative forcing anomaly than eruptions in 1458, 1641, 1809, and
1815 CE, let alone the large Medieval eruption of Samalas (1257 CE)
(Verosub and Lippman, 2008; Lavigne et al., 2013; Sigl et al., 2015).
Our finding here of a large-scale, coherent cooling in response to
explosive volcanism is yet further evidence (Anchukaitis et al.,
2012; Brohan et al., 2012; D'Arrigo et al., 2013; Esper et al.,
2013b,a; St. George et al., 2013; Büntgen et al., 2014; Jull et al.,
2014; Esper et al., 2015; Sigl et al., 2015; Stoffel et al., 2015;
Wilson et al., 2016) against the hypothesis that tree-ring proxies
are missing the volcanic cooling signal due to undetected absent
rings (Mann et al., 2012, 2013).
Low frequency coherence between solar variability and our
reconstruction appears to be a stable characteristic through time
(Figs. 14 and 15), likely linked to reconstructed cold anomalies
during solar Grand Minima. Because variations in total solar irradiance are relatively small, on the order of a few tenths of a W m2,
the mechanism that could result in a detectable cooling remains
uncertain. The most likely connection is via changes in large-scale
Northern Hemisphere circulation, which favor colder temperature
over continents (e.g. Shindell et al., 2001a, 2003; Swingedouw
et al., 2010) and thus would be captured in our reconstruction.
Nevertheless, while variability in solar forcing may be important on
bicentennial and perhaps at continental scales, fingerprinting
suggests that the solar effect in the hemisphere-scale anomalies is
otherwise relatively small and that volcanic forcing is more
important overall in determining pre-industrial temperature trajectories (Schurer et al., 2014; McGregor et al., 2015). There is no
sign in our reconstruction of a discernible temperature response to
the shorter 11 and 22 year sunspot cycle (Schwabe/Hale), which is
consistent with other investigations of the insolation signal in tree
rings (e.g. Briffa, 1994). There are a number of possible reasons for
the absence of this signal. Internal climate system variability is
substantially stronger at interannual and decadal time scales,
which may prevent statistical detection of solar influences with
similar frequencies, but still allow it at the centennial scale when
the magnitude of internal variability is smaller than the forced
signal. Short-term climate anomalies caused by explosive volcanism could also disrupt detection of a decadal solar signal. The
temperature response to solar variability at lower frequencies may
also reflect slow temperature feedbacks that enhance its direct
effect.
Over those grid points available back to 950 CE with minimum
level of reconstruction skill (RE > 0), ~90% show warmer conditions
during the MCA than during the LIA, with a field median difference
of 0.32 C. Removing likely individual grid point outliers (Greenland
and northeastern North America, see above) results in a slightly
smaller epochal field median difference (0.30 C) and a range of grid
point values of 0.64 to þ1.05 C. Mann et al. (2009) calculated a
0.24 C global summer mean difference between MCA and LIA, but
the difference in season, spatial domain and geographic extent, and
the ‘fragility’ (Wang et al., 2015) of reconstructing a cold Medieval
tropical Pacific make any direct comparison difficult. Calculating
the MCA-LIA epochal difference using the spatial mean time series
(Fig. 12) gives a value of 0.36 C, approximately in the middle of the
distribution for other large-scale Northern Hemisphere reconstructions, and within the higher end of the range of values
ndez-Donado et al., 2013;
from climate model simulations (Ferna
Wilson et al., 2016).
17
5. Conclusions and future work
We have reconstructed the extratropical Northern Hemisphere
MJJA temperature anomaly field back to 750 CE using a network of
temperature-sensitive predictors. The reconstruction shows significant field skill associated with proximity to the predictors,
particularly where proxy density data are available. In other words,
we observe the highest reconstruction skill and smallest errors
where we have the most sensitive tree-ring proxies, whereas larger
errors and lower skill are associated with grid points distal from the
predictor network or where only tree-ring width data are available,
particularly in North America. These observations will be used to
guide future sampling and proxy development priorities, including
the development of new sites, efforts to increase the number of
MXD and BI series, and the extension in time of existing high
quality chronologies.
Our field reconstruction reveals coherent responses to
changes in radiative forcing over the last 1200 years, including
the influence of solar and volcanic forcing. Future research with
our field reconstruction will use fingerprint detection (Hegerl
et al., 2007; Schurer et al., 2013, 2014) to quantitatively assess
the role of forcing and internal variability, including identification of spatial patterns linked to large-scale modes of variability
and specific forcing agents. Formal, quantitative comparison between our reconstruction and paleoclimate model simulations
(Schmidt et al., 2012; Kageyama et al., 2016) will be used to
assess climate model performance and to investigate the
dynamical context for reconstructed spatial temperature anomalies. Using proxy system models (Evans et al., 2013), the NTREND network could also be applied within an offline data
assimilation framework (Steiger et al., 2014; Hakim et al., 2016).
Finally, our spatially-explicit reconstructions can be used to
explore and understand the possible role of past temperature
variability e especially due to volcanic eruptions e in contributing to historical societal dynamics, resilience, and change
(McCormick et al., 2007; Ludlow et al., 2013; Sigl et al., 2015;
Büntgen et al., 2016).
Acknowledgements
The N-TREND consortium is not itself funded, but many individuals acknowledge relevant projects, grants, and support; KJA:
National Science Foundation Paleoclimate Perspectives on Climate
Change NSF AGS-1501856 and NSF AGS-1501834; RW: UK Natural
Environment Research Council (NERC - NE/K003097/1) and Leverhulme Trust project (F/00 268/BG); KRB and TO: NERC (Belmont
Forum/JPI-Climate: INTEGRATE project NE/P006809/1); KRB also
thanks Gaurav Kapur FRCR and Colin Watts FRCS for time; UB:
Czech project ‘Building up a multidisciplinary scientific team
focused on drought’ No. CZ.1.07/2.3.00/20.0248; EC: National Science Foundation Paleoclimate Perspectives on Climate Change
AGS-1502224; RD: National Science Foundation Paleoclimate Perspectives on Climate Change AGS-1159430, AGS-1502150, and AGS1502224; SH: Academy of Finland; GH and AS: ERC advanced grant
TITAN (EC-320691), NCAS GH specifically with a Royal Society
Wolfson Research Merit Award (WM130060); GH and AS: PACMEDY, NE/P006752/1; HL: The Swedish Science Council (VR) (20125246); MR: The Carnegie Trust for the Universities of Scotland; GW:
NSF AGS-1502186. The N-TREND project website, along with the
archived TR chronologies and temperature reconstructions can be
found
at
https://www.ncdc.noaa.gov/cdo/f?p¼519:1:0::::P1_
study_id:19743 and additional information is available at https://
ntrenddendro.wordpress.com/. Lamont-Doherty Earth Observatory contribution #8093.
18
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
References
Ammann, C.M., Joos, F., Schimel, D.S., Otto-Bliesner, B.L., Tomas, R.A., 2007. Solar
influence on climate during the past millennium: results from transient simulations with the NCAR Climate System Model. Proc. U. S. Natl. Acad. Sci. 104,
3713e3718.
Ammann, C.M., Naveau, P., 2003. Statistical analysis of tropical explosive volcanism
occurrences over the last 6 centuries. Geophys. Res. Lett. 30, 1210. http://
dx.doi.org/10.1029/2002GL016388.
Anchukaitis, K., McKay, N., 2014. PAGES2k: advances in climate field reconstructions. PAGES Mag. 22, 98.
Anchukaitis, K.J., Breitenmoser, P., Briffa, K.R., Buchwal, A., Büntgen, U., Cook, E.R.,
D'Arrigo, R.D., Esper, J., Evans, M.N., Frank, D., Grudd, H., Gunnarson, B.,
€ rner, C., Krusic, P.J., Luckman, B., Melvin, T.M.,
Hughes, M.K., Kirdyanov, A.V., Ko
Salzer, M.W., Shashkin, A.V., Timmreck, C., Vaganov, E.A., Wilson, R.J., 2012. Tree
rings and volcanic cooling. Nat. Geosci. 5, 836e837. http://dx.doi.org/10.1038/
ngeo1645.
Anchukaitis, K.J., Buckley, B.M., Cook, E.R., Cook, B.I., D'Arrigo, R., Ammann, C.M.,
2010. The influence of volcanic eruptions on the climate of the Asian monsoon
region. Geophys. Res. Lett. 37 http://dx.doi.org/10.1029/2010GL044843.
Anchukaitis, K.J., D'Arrigo, R.D., Andreu-Hayles, L., Frank, D., Verstege, A., Curtis, A.,
Buckley, B.M., Jacoby, G.C., Cook, E.R., 2013. Tree-ring-reconstructed summer
temperatures from northwestern North America during the last nine centuries.
J. Clim. 26, 3001e3012. http://dx.doi.org/10.1175/JCLI-D-11-00139.1.
Anchukaitis, K.J., Evans, M.N., Kaplan, A., Vaganov, E.A., Hughes, M.K., GrissinoMayer, H.D., Cane, M.A., 2006. Forward modeling of regional scale tree-ring
patterns in the southeastern United States and the recent influence of summer drought. Geophys. Res. Lett. 33 http://dx.doi.org/10.1029/2005GL025050.
Andreu-Hayles, L., D'Arrigo, R.D., Anchukaitis, K.J., Beck, P., Frank, D., Verstege, A.,
Goetz, S., 2011. Varying boreal forest response to Arctic environmental change
at the Firth River, Alaska. Environ. Res. Lett. 6, 045503. http://dx.doi.org/
10.1088/1748-9326/6/4/045503.
Andronova, N.G., Schlesinger, M.E., 2000. Causes of global temperature changes
during the 19th and 20th centuries. Geophys. Res. Lett. 27, 2137e2140.
gin, Y., 2015. Biases in RCS tree ring
Autin, J., Gennaretti, F., Arseneault, D., Be
chronologies due to sampling heights of trees. Dendrochronologia 36, 13e22.
Biondi, F., Perkins, D.L., Cayan, D.R., Hughes, M.K., 1999. July temperature during the
second millennium reconstructed from Idaho tree rings. Geophys. Res. Lett. 26,
1445e1448.
€ rklund, J., Gunnarson, B.E., Seftigen, K., Zhang, P., Linderholm, H.W., 2015. Using
Bjo
adjusted Blue Intensity data to attain high-quality summer temperature information: a case study from Central Scandinavia. Holocene 25, 547e556.
€ rklund, J.A., Gunnarson, B.E., Seftigen, K., Esper, J., Linderholm, H.W., 2014. Blue
Bjo
intensity and density from northern Fennoscandian tree rings, exploring the
potential to improve summer temperature reconstructions with earlywood
information. Clim. Past 10, 877e885. http://dx.doi.org/10.5194/cp-10-877-2014.
Bowman, A.W., Azzalini, A., 1997. Applied Smoothing Techniques for Data Analysis.
Oxford University Press.
Bradley, R.S., Briffa, K.R., Crowley, T.J., Hughes, M.K., Jones, P.D., Mann, M.E., 2001.
The scope of medieval warming. Science 292, 2011e2012.
€nnimann, S., Frank, D., Steinhilber, F., Wanner, H., 2012.
Breitenmoser, P., Beer, J., Bro
Solar and volcanic fingerprints in tree-ring chronologies over the past 2000
years. Palaeogeogr. Palaeoclim. Palaeoecol. 313e314, 127e139.
Briffa, K., Jones, P., 1993. Global surface air temperature variations during the
twentieth century: Part 2, implications for large-scale high-frequency palaeoclimatic studies. Holocene 3, 77e88.
Briffa, K.R., 1994. Grasping at shadows? a selective review of the search for sunspotrelated variability in tree rings, in: the solar engine and its influence on
terrestrial atmosphere and climate. Springer Nat. 417e435. http://dx.doi.org/
10.1007/978-3-642-79257-1_26.
Briffa, K.R., 1995. Interpreting high-resolution proxy climate data e the example of
dendroclimatology. In: von Storch, H., Navarra, A. (Eds.), Analysis of Climate
Variability: Applications of Statistical Techniques. Springer, pp. 77e94. http://
dx.doi.org/10.1007/978-3-662-03167-4_5.
Briffa, K.R., 2000. Annual climate variability in the Holocene: interpreting the
message of ancient trees. Quat. Sci. Rev. 19, 87e105.
n, W., Shiyatov, S.G., 1996. TreeBriffa, K.R., Jones, P.D., Schweingruber, F.H., Karle
ring variables as proxy-climate indicators: problems with low-frequency signals, in: climatic variations and forcing mechanisms of the last 2000 years.
Springer Nat. 9e41. http://dx.doi.org/10.1007/978-3-642-61113-1_2. URL:
https://doi.org/10.1007%2F978-3-642-61113-1_2.
Briffa, K.R., Jones, P.D., Schweingruber, F.H., Osborn, T.J., 1998a. Influence of volcanic
eruptions on Northern Hemisphere summer temperature over the past 600
years. Nature 393, 450e455.
Briffa, K.R., Melvin, T.M., 2011. A closer look at regional curve standardization of
tree-ring records: justification of the need, a warning of some pitfalls, and
suggested improvements in its application. In: Hughes, M.K., Diaz, H.F.,
Swetnam, T.W. (Eds.), Dendroclimatology: Progress and Prospects. Springer
Verlag, pp. 113e146 volume 11 of Developments in Paleoenvironmental
Research.
Briffa, K.R., Melvin, T.M., Osborn, T.J., Hantemirov, R.M., Kirdyanov, A.V.,
Mazepa, V.S., Shiyatov, S.G., Esper, J., 2013. Reassessing the evidence for treegrowth and inferred temperature change during the Common Era in Yamalia,
northwest Siberia. Quat. Sci. Rev. 72, 83e107. http://dx.doi.org/10.1016/
j.quascirev.2013.04.008.
Briffa, K.R., Osborn, T.J., Schweingruber, F.H., 2004. Large-scale temperature inferences from tree rings: a review. Glob. Planet. Change 40, 11e26.
Briffa, K.R., Osborn, T.J., Schweingruber, F.H., Jones, P.D., Shiyatov, S.G., Vaganov, E.A.,
2002. Tree-ring width and density data around the Northern Hemisphere: Part
1, local and regional climate signals. Holocene 12, 737e757.
Briffa, K.R., Schweingruber, F.H., Jones, P.D., Osborn, T.J., Harris, I.C., Shiyatov, S.G.,
Vaganov, E.A., Grudd, H., 1998b. Trees tell of past climates: but are they
speaking less clearly today? Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 353,
65e73.
Brohan, P., Allan, R., Freeman, E., Wheeler, D., Wilkinson, C., Williamson, F., 2012.
Constraining the temperature history of the past millennium using early
instrumental observations. Clim. Past 8, 1551e1563. http://dx.doi.org/10.5194/
cp-8-1551-2012.
Buckley, B., Anchukaitis, K., Penny, D., Fletcher, R., Cook, E., Sano, M., Nam, L.,
Wichienkeeo, A., Minh, T., Hong, T., 2010. Climate as a contributing factor in the
demise of Angkor, Cambodia. Proc. Natl. Acad. Sci. 107, 6748e6752.
Büntgen, U., Frank, D.C., Nievergelt, D., Esper, J., 2006. Summer temperature variations in the European Alps, A.D.. 755e2004. J. Clim. 19, 5606e5623. http://
dx.doi.org/10.1175/jcli3917.1.
Büntgen, U., Kyncl, T., Ginzler, C., Jacks, D.S., Esper, J., Tegel, W., Heussner, K.U.,
Kyncl, J., 2013. Filling the Eastern European gap in millennium-long temperature reconstructions. Proc. Natl. Acad. Sci. 110, 1773e1778. http://dx.doi.org/
10.1073/pnas.1211485110.
Büntgen, U., Myglan, V.S., Ljungqvist, F.C., McCormick, M., Cosmo, N.D., Sigl, M.,
Jungclaus, J., Wagner, S., Krusic, P.J., Esper, J., Kaplan, J.O., de Vaan, M.A.C.,
Luterbacher, J., Wacker, L., Tegel, W., Kirdyanov, A.V., 2016. Cooling and societal
change during the late antique little ice age from 536 to around 660 AD. Nat.
Geosci. 9, 231e236. http://dx.doi.org/10.1038/ngeo2652.
Büntgen, U., Tegel, W., Nicolussi, K., McCormick, M., Frank, D., Trouet, V., Kaplan, J.,
Herzig, F., Heussner, K., Wanner, H., Luterbacher, J., Esper, J., 2011. 2500 years of
European climate variability and human susceptibility. Science 331, 578e582.
Büntgen, U., Wacker, L., Nicolussi, K., Sigl, M., Güttler, D., Tegel, W., Krusic, P.J.,
Esper, J., 2014. Extraterrestrial confirmation of tree-ring dating. Nat. Clim.
Change 4, 404e405.
Campbell, R., McCarroll, D., Loader, N.J., Grudd, H., Robertson, I., Jalkanen, R., 2007.
Blue intensity in Pinus sylvestris tree-rings: developing a new palaeoclimate
proxy. Holocene 17, 821e828.
Cole-Dai, J., Ferris, D.G., Lanciki, A.L., Savarino, J., Thiemens, M.H., McConnell, J.R.,
2013. Two likely stratospheric volcanic eruptions in the 1450s C.E. found in a
bipolar, subannually dated 800 year ice core record. J. Geophys. Res. Atmos. 118,
7459e7466. http://dx.doi.org/10.1002/jgrd.50587.
kstis, L. (Eds.), 1990. Methods of Dendrochronology: Applications in
Cook, E., Kairiu
the Environmental Sciences. Kluwer Academic, Doredrecht.
Cook, E.R., 1987. The decomposition of tree-ring series for environmental studies.
Tree-Ring Bull. 47, 37e59.
Cook, E.R., Anchukaitis, K.J., Buckley, B.M., D'Arrigo, R.D., Jacoby, G.C., Wright, W.E.,
2010a. Asian monsoon failure and megadrought during the last millennium.
Science 328, 486e489. http://dx.doi.org/10.1126/science.1185188.
Cook, E.R., Briffa, K.R., Jones, P.D., 1994. Spatial regression methods in dendroclimatology e a review and comparison of 2 techniques. Int. J. Climatol. 14,
379e402.
Cook, E.R., Briffa, K.R., Meko, D.M., Graybill, D.A., Funkhouser, G., 1995. The Segment
Length Curse in long tree-ring chronology development for paleoclimatic
studies. Holocene 5, 229e237.
Cook, E.R., Krusic, P.J., Anchukaitis, K.J., Buckley, B.M., Nakatsuka, T., Sano, M., 2013.
Tree-ring reconstructed summer temperature anomalies for temperate East
Asia since 800 CE. Clim. Dynam 41, 2957e2972.
Cook, E.R., Meko, D.M., Stahle, D.W., Cleaveland, M.K., 1999. Drought reconstructions for the continental United States. J. Clim. 12, 1145e1162.
Cook, E.R., Seager, R., Heim, R.R., Vose, R.S., Herweijer, C., Woodhouse, C.A., 2010b.
Megadroughts in North America: placing IPCC projections of hydroclimatic
change in a long-term paleoclimate context. J. Quat. Sci. 25, 48e61.
Cowtan, K., Way, R.G., 2014. Coverage bias in the HadCRUT4 temperature series and
its impact on recent temperature trends. Q. J. R. Meteorol. Soc. 140, 1935e1944.
http://dx.doi.org/10.1002/qj.2297.
Crowley, T.J., 2000. Causes of climate change over the past 1000 years. Science 289,
270e277.
Crowley, T.J., Zielinski, G., Vinther, B., Udisti, R., Kreutz, K., Cole-Dai, J., Castellano, E.,
2008. Volcanism and the little ice age. PAGES News 16, 22e23.
D'Arrigo, R., Anchukaitis, K.J., Buckley, B., Cook, E.R., Wilson, R., 2011. Regional climatic and North Atlantic Oscillation signatures in West Virginia red cedar over
the past millennium. Glob. Planet. Change 84e85, 8e13. http://dx.doi.org/
10.1016/j.gloplacha.2011.07.003.
D'Arrigo, R., Buckley, B., Kaplan, S., Woollett, J., 2003. Interannual to multidecadal
modes of Labrador climate variability inferred from tree rings. Clim. Dynam 20,
219e228.
D'Arrigo, R., Jacoby, G., Free, R., 1992. Tree-ring width and maximum latewood
density at the North American tree line: parameters of climatic change. Can. J.
For. Res. 22, 1290e1296.
D'Arrigo, R., Mashig, E., Frank, D., Jacoby, G., Wilson, R., 2004. Reconstructed warm
season temperatures for Nome, Seward peninsula, Alaska. Geophys. Res. Lett. 31
http://dx.doi.org/10.1029/2004gl019756.
D'Arrigo, R., Wilson, R., Anchukaitis, K.J., 2013. Volcanic cooling signal in tree ring
temperature records for the past millennium. J. Geophys. Res. Atmos. 118,
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
9000e9010. http://dx.doi.org/10.1002/jgrd.50692.
D'Arrigo, R., Wilson, R., Liepert, B., Cherubini, P., 2008. On the ‘Divergence Problem’
in Northern Forests: a review of the tree-ring evidence and possible causes.
Glob.
Planet.
Change
60,
289e305.
http://dx.doi.org/10.1016/
j.gloplacha.2007.03.004.
D'Arrigo, R., Wilson, R., Wiles, G., Anchukaitis, K., Solomina, O., Davi, N., Deser, C.,
Dolgova, E., 2015. Tree-ring reconstructed temperature index for coastal
northern Japan: implications for western North Pacific variability. Int. J. Climatol. 35, 3713e3720.
D'Arrigo, R.D., Jacoby, G.C., Buckley, B.M., Sakulich, J., Frank, D., Wilson, R., Curtis, A.,
Anchukaitis, K.J., 2009. Tree growth and inferred temperature variability at the
North American Arctic treeline. Glob. Planet. Change 65, 71e82.
D'Arrigo, R.D., Wilson, R., Jacoby, G.C., 2006. On the long-term context for late
twentieth century warming. J. Geophys. Res. 111 http://dx.doi.org/10.1029/
2005JD006352.
Davi, N., D'Arrigo, R., Jacoby, G., Cook, E., Anchukaitis, K., Nachin, B., Rao, M.,
Leland, C., 2015. A long-term context (931e2005 C.E.) for rapid warming over
Central Asia. Quat. Sci. Rev. 121, 89e97. http://dx.doi.org/10.1016/
j.quascirev.2015.05.020.
Davi, N.K., Jacoby, G.C., Wiles, G.C., 2003. Boreal temperature variability inferred
from maximum latewood density and tree-ring width data, Wrangell Mountain
region, Alaska. Quat. Res. 60, 252e262.
Delaygue, G., Bard, E., 2011. An antarctic view of beryllium-10 and solar activity for
the past millennium. Clim. Dynam 36, 2201e2218. http://dx.doi.org/10.1007/
s00382-010-0795-1.
~a
n, I., Büntgen, U., Gonz
mezDorado-Lin
alez-Rouco, F., Zorita, E., Mont
avez, J.P., Go
rrez, E., 2012. Estimating 750
Navarro, J.J., Brunet, M., Heinrich, I., Helle, G., Gutie
years of temperature variations and uncertainties in the Pyrenees by tree-ring
reconstructions and climate simulations. Clim. Past 8, 919e933. http://
dx.doi.org/10.5194/cp-8-919-2012.
Edwards, T.L., Crucifix, M., Harrison, S.P., 2007. Using the past to constrain the
future: how the palaeorecord can improve estimates of global warming. Prog.
Phys. Geogr. 31, 481e500.
Emile-Geay, J., Cobb, K.M., Mann, M.E., Wittenberg, A.T., 2013. Estimating central
equatorial Pacific SST variability over the past millennium. Part II: reconstructions and implications. J. Clim. 26, 2329e2352.
Esper, J., Büntgen, U., Luterbacher, J., Krusic, P.J., 2013a. Testing the hypothesis of
post-volcanic missing rings in temperature sensitive dendrochronological data.
Dendrochronologia
31,
216e222.
http://dx.doi.org/10.1016/
j.dendro.2012.11.002.
Esper, J., Cook, E.R., Schweingruber, F.H., 2002. Low-frequency signals in long treering chronologies for reconstructing past temperature variability. Science 295,
2250e2253.
Esper, J., Düthorn, E., Krusic, P.J., Timonen, M., Büntgen, U., 2014. Northern European
summer temperature variations over the Common Era from integrated treering density records. J. Quat. Sci. 29, 487e494. http://dx.doi.org/10.1002/
jqs.2726.
Esper, J., Frank, D., Büntgen, U., Verstege, A., Hantemirov, R.M., Kirdyanov, A.V., 2010.
Trends and uncertainties in Siberian indicators of 20th century warming. Glob.
Change Biol. 16, 386e398.
Esper, J., Frank, D., Büntgen, U., Verstege, A., Luterbacher, J., Xoplaki, E., 2007. Longterm drought severity variations in Morocco. Geophys. Res. Lett. 34 http://
dx.doi.org/10.1029/2007GL030844.
Esper, J., Frank, D.C., Timonen, M., Zorita, E., Wilson, R.J.S., Luterbacher, J.,
Holzkamper, S., Fischer, N., Wagner, S., Nievergelt, D., Verstege, A., Buntgen, U.,
2012. Orbital forcing of tree-ring data. Nat. Clim. Change 2, 862e866. http://
dx.doi.org/10.1038/nclimate1589.
Esper, J., Krusic, P.J., Ljungqvist, F.C., Luterbacher, J., Carrer, M., Cook, E., Davi, N.K.,
Hartl-Meier, C., Kirdyanov, A., Konter, O., et al., 2016. Ranking of tree-ring based
temperature reconstructions of the past millennium. Quat. Sci. Rev. 145,
134e151.
Esper, J., Schneider, L., Krusic, P.J., Luterbacher, J., Büntgen, U., Timonen, M.,
Sirocko, F., Zorita, E., 2013b. European summer temperature response to
annually dated volcanic eruptions over the past nine centuries. Bull. Volcanol.
75 http://dx.doi.org/10.1007/s00445-013-0736-z.
€ne, B.R., Büntgen, U., 2015. Signals and
Esper, J., Schneider, L., Smerdon, J.E., Scho
memory in tree-ring width and density data. Dendrochronologia 35, 62e70.
Esper, J., Wilson, R.J.S., Frank, D.C., Moberg, A., Wanner, H., Luterbacher, J., 2005.
Climate: past ranges and future changes. Quat. Sci. Rev. 24, 2164e2166.
pez, B., 2013. Statistically derived contributions of
Estrada, F., Perron, P., Martínez-Lo
diverse human influences to twentieth-century temperature changes. Nat.
Geosci. 6, 1050e1055.
Evans, M., Tolwinski-Ward, S., Thompson, D., Anchukaitis, K., 2013. Applications of
proxy system modeling in high resolution paleoclimatology. Quat. Sci. Rev. 76,
16e28. http://dx.doi.org/10.1016/j.quascirev.2013.05.024.
Evans, M.N., Kaplan, A., Cane, M.A., 2002. Pacific sea surface temperature field
reconstruction from coral d1O data using reduced space objective analysis.
Paleoceanography 17. http://dx.doi.org/10.1029/2000PA000590.
Evans, M.N., Kaplan, A., Cane, M.A., Villalba, R., 2001. Globality and optimality in
climate field reconstructions from proxy data. In: Markgraf, V. (Ed.), Interhemispheric Climate Linkages. Cambridge University Press, Cambridge, UK,
pp. 53e72.
ndez-Donado, L., Gonza
lez-Rouco, J.F., Raible, C.C., Ammann, C.M.,
Ferna
Barriopedro, D., García-Bustamante, E., Jungclaus, J.H., Lorenz, S.J.,
Luterbacher, J., Phipps, S.J., Servonnat, J., Swingedouw, D., Tett, S.F.B., Wagner, S.,
19
Yiou, P., Zorita, E., 2013. Large-scale temperature response to external forcing in
simulations and reconstructions of the last millennium. Clim. Past 9, 393e421.
http://dx.doi.org/10.5194/cp-9-393-2013.
€ hm, R., Maugeri, M., Esper, J., 2007. Warmer early instruFrank, D., Büntgen, U., Bo
mental measurements versus colder reconstructed temperatures: shooting at a
moving target. Quat. Sci. Rev. 26, 3298e3310.
Frank, D., Esper, J., Zorita, E., Wilson, R., 2010. A noodle, hockey stick, and spaghetti
plate: a perspective on high-resolution paleoclimatology. WIREs Clim. Chg 1,
507e516. http://dx.doi.org/10.1002/wcc.53.
Fritts, H.C., 1976. Tree Rings and Climate. Academic Press, New York.
Fritts, H.C., 1991. Reconstructing Large-scale Climatic Patterns from Tree-ring Data:
a Diagnostic Analysis. University of Arizona Press, London.
Gallant, A.J., Phipps, S.J., Karoly, D.J., Mullan, A.B., Lorrey, A.M., 2013. Nonstationary
Australasian teleconnections and implications for paleoclimate reconstructions.
J. Clim. 26, 8827e8849.
Gao, C., Robock, A., Ammann, C.M., 2008. Volcanic forcing of climate over the past
1500 years: an improved ice-core-based index for climate models. J. Geophys.
Res. 113 http://dx.doi.org/10.1029/2008JD010239.
Gao, C.C., Robock, A., Self, S., Witter, J.B., Steffenson, J.P., Clausen, H.B., SiggaardAndersen, M.L., Johnsen, S., Mayewski, P.A., Ammann, C., 2006. The 1452 or
1453 AD Kuwae eruption signal derived from multiple ice core records: greatest
volcanic sulfate event of the past 700 years. J. Geophys. Res. - Atmos. 111 http://
dx.doi.org/10.1029/2005JD006710.
gin, Y., 2014. VolcanoGennaretti, F., Arseneault, D., Nicault, A., Perreault, L., Be
induced regime shifts in millennial tree-ring chronologies from northeastern
North America. Proc. Natl. Acad. Sci. 111, 10077e10082.
Gershunov, A., Barnett, T.P., 1998. Interdecadal modulation of ENSO teleconnections.
Bull. Am. Meteorol. Soc. 79, 2715e2725.
Gillett, N.P., Arora, V.K., Flato, G.M., Scinocca, J.F., von Salzen, K., 2012. Improved
constraints on 21st-century warming derived using 160 years of temperature
observations. Geophys. Res. Lett. 39 http://dx.doi.org/10.1029/2011gl050226.
Goosse, H., 2016. Reconstructed and simulated temperature asymmetry between
continents in both hemispheres over the last centuries. Clim. Dyn. http://
dx.doi.org/10.1007/s00382-016-3154-z.
Grinsted, A., Moore, J.C., Jevrejeva, S., 2004. Application of the cross wavelet
transform and wavelet coherence to geophysical time series. Nonlinear Process.
Geophys. 11, 561e566.
Hakim, G.J., Emile-Geay, J., Steig, E.J., Noone, D., Anderson, D.M., Tardif, R., Steiger, N.,
Perkins, W.A., 2016. The last millennium climate reanalysis project: framework
and first results. J. Geophys. Res. Atmos. 121, 6745e6764. http://dx.doi.org/
10.1002/2016jd024751.
Harris, I., Jones, P.D., Osborn, T.J., Lister, D.H., 2013. Updated high-resolution grids of
monthly climatic observations: the CRU TS3.10 dataset. Int. J. Climatol. http://
dx.doi.org/10.1002/joc.3711.
Harrison, S., Bartlein, P., Izumi, K., Li, G., Annan, J., Hargreaves, J., Braconnot, P.,
Kageyama, M., 2015. Evaluation of CMIP5 palaeo-simulations to improve
climate projections. Nat. Clim. Change 5, 735e743.
Haurwitz, M.W., Brier, G.W., 1981. A critique of the superposed epoch analysis
method- its application to solar-weather relations. Mon. Weather Rev. 109,
2074e2079.
Hegerl, G., Stott, P., 2014. From past to future warming. Science 343, 844e845.
Hegerl, G.C., Crowley, T.J., Allen, M., Hyde, W.T., Pollack, H.N., Smerdon, J., Zorita, E.,
2007. Detection of human influence on a new, validated 1500-year temperature
reconstruction. J. Clim. 20, 650e666.
Hegerl, G.C., Crowley, T.J., Baum, S.K., Kim, K.Y., Hyde, W.T., 2003. Detection of
volcanic, solar and greenhouse gas signals in paleo-reconstructions of Northern
Hemispheric temperature. Geophys. Res. Lett. 30 http://dx.doi.org/10.1029/
2002GL016635.
Hegerl, G.C., Hasselmann, K., Cubasch, U., Mitchell, J., Roeckner, E., Voss, R.,
Waszkewitz, J., 1997. Multi-fingerprint detection and attribution analysis of
greenhouse gas, greenhouse gas-plus-aerosol and solar forced climate change.
Clim. Dyn. 13, 613e634.
Hegerl, G.C., Karl, T.R., Allen, M., Bindoff, N.L., Gillett, N., Karoly, D., Zhang, X.,
Zwiers, F., 2006. Climate change detection and attribution: beyond mean
temperature signals. J. Clim. 19, 5058e5077. http://dx.doi.org/10.1175/jcli3900.1.
Hegerl, G.C., Russon, T., 2011. Using the past to predict the future? Science 334,
1360e1361.
€ m, T., Meril€
Helama, S., Vartiainen, M., Holopainen, J., M€
akel€
a, H., Kolstro
ainen, J.,
2014. A palaeotemperature record for the Finnish Lakeland based on microdensitometric variations in tree rings. Geochronometria 41. http://dx.doi.org/
10.2478/s13386-013-0163-0.
Hughes, M., 2002. Dendrochronology in climatology - the state of the art. Dendrochronologia 20, 95e116.
Hughes, M.K., Diaz, H.F., 1994. Was there a ‘Medieval Warm Period’ and if so. Clim.
Change 26, 109e142.
Hughes, M.K., Vaganov, E.A., Shiyatov, S., Touchan, R., Funkhouser, G., 1999. Twentieth-century summer warmth in northern Yakutia in a 600-year context. Holocene 9, 629e634.
Jacoby, G.C., D'Arrigo, R.D., 1995. Tree ring width and density evidence of climatic
and potential forest change in Alaska. Glob. Biogeochem. Cycles 9, 227e234.
Jacoby, G.C., Lovelius, N.V., Shumilov, O.I., Raspopov, O.M., Karbainov, J.M.,
Frank, D.C., 2000. Long-term temperature trends and tree growth in the Taymir
region of northern Siberia. Quat. Res. 53, 312e318.
Jones, G.S., Stott, P.A., Christidis, N., 2013. Attribution of observed historical
nearesurface temperature variations to anthropogenic and natural causes using
20
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
CMIP5 simulations. J. Geophys. Res. Atmos. 118, 4001e4024.
Jones, P.D., Briffa, K.R., Osborn, T.J., Lough, J.M., van Ommen, T.D., Vinther, B.M.,
Luterbacher, J., Wahl, E.R., Zwiers, F.W., Mann, M.E., Schmidt, G.A.,
Ammann, C.M., Buckley, B.M., Cobb, K.M., Esper, J., Goosse, H., Graham, N.,
Jansen, E., Kiefer, T., Kull, C., Küttel, M., Mosley-Thompson, E., Overpeck, J.T.,
Riedwyl, N., Schulz, M., Tudhope, A.W., Villalba, R., Wanner, H., Wolff, E.,
Xoplaki, E., 2009. High-resolution palaeoclimatology of the last millennium: a
review of current status and future prospects. Holocene 19, 3e49. http://
dx.doi.org/10.1177/0959683608098952.
Jones, P.D., Lister, D.H., Osborn, T.J., Harpham, C., Salmon, M., Morice, C.P., 2012.
Hemispheric and large-scale land-surface air temperature variations: an
extensive revision and an update to 2010. J. Geophys. Res. Atmos. 117 http://
dx.doi.org/10.1029/2011jd017139.
Jones, P.D., Osborn, T.J., Briffa, K.R., 1997. Estimating sampling errors in large-scale
temperature averages. J. Clim. 10, 2548e2568.
Jull, A.J.T., Panyushkina, I.P., Lange, T.E., Kukarskih, V.V., Myglan, V.S., Clark, K.J.,
Salzer, M.W., Burr, G.S., Leavitt, S.W., 2014. Excursions in the 14 C record at A.D.
774-775 in tree rings from Russia and America. Geophys. Res. Lett. 41,
3004e3010. http://dx.doi.org/10.1002/2014gl059874.
Kageyama, M., Braconnot, P., Harrison, S.P., Haywood, A.M., Jungclaus, J., OttoBliesner, B.L., Peterschmitt, J.Y., Abe-Ouchi, A., Albani, S., Bartlein, P.J.,
Brierley, C., Crucifix, M., Dolan, A., Fernandez-Donado, L., Fischer, H.,
Hopcroft, P.O., Ivanovic, R.F., Lambert, F., Lunt, D.J., Mahowald, N.M.,
Peltier, W.R., Phipps, S.J., Roche, D.M., Schmidt, G.A., Tarasov, L., Valdes, P.J.,
Zhang, Q., Zhou, T., 2016. PMIP4-CMIP6: the contribution of the paleoclimate
modelling intercomparison project to CMIP6. Geosci. Model Dev. Discuss.
http://dx.doi.org/10.5194/gmd-2016-106.
Kaplan, J.O., Krumhardt, K.M., Ellis, E.C., Ruddiman, W.F., Lemmen, C.,
Goldewijk, K.K., 2011. Holocene carbon emissions as a result of anthropogenic
land cover change. Holocene 21, 775e791. http://dx.doi.org/10.1177/
0959683610386983.
Kaufman, D., 2014. A community-driven framework for climate reconstructions.
Eos. Trans. Am. Geophys. Union 95, 361e362.
Keigwin, L.D., Sachs, J.P., Rosenthal, Y., 2003. A 1600-year history of the Labrador
current off Nova Scotia. Clim. Dyn. 21, 53e62. http://dx.doi.org/10.1007/s00382003-0316-6.
Kennedy, J.J., Rayner, N.A., Smith, R.O., Parker, D.E., Saunby, M., 2011a. Reassessing
biases and other uncertainties in sea surface temperature observations
measured in situ since 1850: 1. measurement and sampling uncertainties.
J. Geophys. Res. 116 http://dx.doi.org/10.1029/2010jd015218.
Kennedy, J.J., Rayner, N.A., Smith, R.O., Parker, D.E., Saunby, M., 2011b. Reassessing
biases and other uncertainties in sea surface temperature observations
measured in situ since 1850: 2. biases and homogenization. J. Geophys. Res. 116
http://dx.doi.org/10.1029/2010jd015220.
Klesse, S., Ziehmer, M., Rousakis, G., Trouet, V., Frank, D., 2015. Synoptic drivers of
400 years of summer temperature and precipitation variability on Mt. Olympus,
Greece. Clim. Dynam 45, 807e824. http://dx.doi.org/10.1007/s00382-014-23133.
Lavigne, F., Degeai, J.P., Komorowski, J.C., Guillet, S., Robert, V., Lahitte, P.,
Oppenheimer, C., Stoffel, M., Vidal, C.M., Surono, Pratomo I., Wassmer, P.,
Hajdas, I., Hadmoko, D.S., de Belizal, E., 2013. Source of the great A.D. 1257
mystery eruption unveiled, Samalas volcano, Rinjani Volcanic Complex,
Indonesia. Proc. Natl. Acad. Sci. 110, 16742e16747. http://dx.doi.org/10.1073/
pnas.1307520110.
Lean, J.L., Rind, D.H., 2008. How natural and anthropogenic influences alter global
and regional surface temperatures: 1889 to 2006. Geophys. Res. Lett. 35 http://
dx.doi.org/10.1029/2008gl034864.
Lee, T.C.K., Zwiers, F.W., Tsao, M., 2008. Evaluation of proxy-based millennial
reconstruction methods. Clim. Dyn. 31, 263e281.
Lehner, F., Raible, C.C., Stocker, T.F., 2012. Testing the robustness of a precipitation
proxy-based North Atlantic Oscillation reconstruction. Quat. Sci. Rev. 45, 85e94.
Lewis, S.C., LeGrande, A.N., 2015. Stability of ENSO and its tropical pacific teleconnections over the last millennium. Clim. Past 11, 1347e1360. http://
dx.doi.org/10.5194/cp-11-1347-2015.
Li, B., Nychka, D., Ammann, C., 2010. The value of multiproxy reconstruction of past
climate. J. Am. Stat. Assoc. 105, 883e895.
€rklund, J., Seftigen, K., Gunnarson, B.E., Fuentes, M., 2015.
Linderholm, H.W., Bjo
Fennoscandia revisited: a spatially improved tree-ring reconstruction of summer temperatures for the last 900 years. Clim. Dynam 45, 933e947. http://
dx.doi.org/10.1007/s00382-014-2328-9.
Luckman, B.H., Wilson, R.J.S., 2005. Summer temperatures in the Canadian Rockies
during the last millennium: a revised record. Clim. Dynam 24, 131e144.
Ludlow, F., Stine, A.R., Leahy, P., Murphy, E., Mayewski, P.A., Taylor, D., Killen, J.,
Baillie, M.G., Hennessy, M., Kiely, G., 2013. Medieval Irish chronicles reveal
persistent volcanic forcing of severe winter cold events, 431e1649 CE. Environ.
Res. Lett. 8, 024035. http://dx.doi.org/10.1088/1748-9326/8/2/024035.
Mann, M.E., Bradley, R.S., Hughes, M.K., 1998. Global-scale temperature patterns
and climate forcing over the past six centuries. Nature 392, 779e787.
Mann, M.E., Fuentes, J.D., Rutherford, S., 2012. Underestimation of volcanic cooling
in tree-ring-based reconstructions of hemispheric temperatures. Nat. Geosci. 5,
202e205.
Mann, M.E., Rutherford, S., 2002. Climate reconstruction using ’pseudoproxies'.
Geophys. Res. Lett. 29 http://dx.doi.org/10.1029/2001GL014554.
Mann, M.E., Rutherford, S., Schurer, A., Tett, S.F., Fuentes, J.D., 2013. Discrepancies
between the modeled and proxy-reconstructed response to volcanic forcing
over the past millennium: implications and possible mechanisms. J. Geophys.
Res. Atmos. 118, 7617e7627.
Mann, M.E., Rutherford, S., Wahl, E., Ammann, C., 2005. Testing the fidelity of
methods used in proxy-based reconstructions of past climate. J. Clim. 18,
4097e4107.
Mann, M.E., Rutherford, S., Wahl, E., Ammann, C., 2007. Robustness of proxy-based
climate field reconstruction methods. J. Geophys. Res. 112 http://dx.doi.org/
10.1029/2006JD008272.
Mann, M.E., Zhang, Z., Hughes, M.K., Bradley, R.S., Miller, S.K., Rutherford, S., Ni, F.,
2008. Proxy-based reconstructions of hemispheric and global surface temperature variations over the past two millennia. Proc. Natl. Acad. Sci. U. S. A. 105,
13252e13257.
Mann, M.E., Zhang, Z., Rutherford, S., Bradley, R.S., Hughes, M.K., Shindell, D.,
Ammann, C., Faluvegi, G., Ni, F., 2009. Global signatures and dynamical origins
of the little ice age and medieval climate anomaly. Science 326, 1256e1260.
lez
Masson-Delmotte, V., Schulz, M., Abe-Ouchi, A., Beer, J., Ganopolski, A., Gonza
Rouco, J., Jansen, E., Lambeck, K., Luterbacher, J., Naish, T., Osborn, T., OttoBliesner, B., Quinn, T., Ramesh, R., Rojas, M., Shao, X., Timmermann, A., 2013.
Information from paleoclimate archives. In: Stocker, T., Qin, D., Plattner, G.K.,
Tignor, M., Allen, S., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P. (Eds.),
Climate Change 2013: the Physical Science Basis. Contribution of Working
Group I to the Fifth Assessment Report of the Intergovernmental Panel on
Climate Change. Cambridge University Press, Cambridge, United Kingdom and
New
York,
NY,
USA,
pp.
383e464.
http://dx.doi.org/10.1017/
CBO9781107415324.013.
Matskovsky, V., Helama, S., 2016. Direct transformation of tree-ring measurements
into palaeoclimate reconstructions in three-dimensional space. Holocene 26,
439e449.
Matskovsky, V.V., Helama, S., 2014. Testing long-term summer temperature
reconstruction based on maximum density chronologies obtained by reanalysis
of tree-ring data sets from northernmost Sweden and Finland. Clim. Past 10,
1473e1487. http://dx.doi.org/10.5194/cp-10-1473-2014.
Matthews, J.A., Briffa, K.R., 2005. The ‘little ice age’: Re-evaluation of an evolving
concept. Geogr. Ann. Ser. A e Phys. Geogr. 87A, 17e36.
McCarroll, D., Loader, N.J., Jalkanen, R., Gagen, M.H., Grudd, H., Gunnarson, B.E.,
Kirchhefer, A.J., Friedrich, M., Linderholm, H.W., Lindholm, M., Boettger, T.,
Los, S.O., Remmele, S., Kononov, Y.M., Yamazaki, Y.H., Young, G.H., Zorita, E.,
2013. A 1200-year multiproxy record of tree growth and summer temperature
at the northern pine forest limit of Europe. Holocene 23, 471e484. http://
dx.doi.org/10.1177/0959683612467483.
McCarroll, D., Pettigrew, E., Luckman, A., Guibal, F., Edouard, J.L., 2002. Blue
reflectance provides a surrogate for latewood density of high-latitude pine tree
rings. Arct. Antarct. Alp. Res. 34, 450e453. http://dx.doi.org/10.2307/1552203.
McCormick, M., Dutton, P.E., Mayewski, P.A., 2007. Volcanoes and the climate
forcing of Carolingian Europe, AD 750e950. Speculum 82, 865e895.
McGregor, H.V., Evans, M.N., Goosse, H., Leduc, G., Martrat, B., Addison, J.A.,
Mortyn, P.G., Oppo, D.W., Seidenkrantz, M.S., Sicre, M.A., Phipps, S.J., Selvaraj, K.,
Thirumalai, K., Filipsson, H.L., Ersek, V., 2015. Robust global ocean cooling trend
for the pre-industrial Common Era. Nat. Geosci. 8, 671e677. http://dx.doi.org/
10.1038/ngeo2510.
McKay, N.P., Kaufman, D.S., 2014. An extended Arctic proxy temperature database
for the past 2,000 years. Sci. Data 1. http://dx.doi.org/10.1038/sdata.2014.26.
Meehl, G.A., Washington, W.M., Ammann, C.M., Arblaster, J.M., Wigley, T.M.L.,
Tebaldi, C., 2004. Combinations of natural and anthropogenic forcings in
twentieth-century climate. J. Clim. 17, 3721e3727.
Meko, D., 1997. Dendroclimatic reconstruction with time varying predictor subsets
of tree indices. J. Clim. 10, 687e696.
Melvin, T., Briffa, K., 2008. A ‘signal-free’ approach to dendroclimatic standardisation. Dendrochronologia 26, 71e86.
Melvin, T.M., 2004. Historical Growth Rates and Changing Climatic Sensitivity of
Boreal Conifers. Ph.D. thesis. University of East Anglia, Norwich, United
Kingdom.
Michaelson, J., 1987. Cross-validation in statistical climate forecast models. J. Clim.
Appl. Meteor 26, 1589e1600.
Morice, C.P., Kennedy, J.J., Rayner, N.A., Jones, P.D., 2012. Quantifying uncertainties
in global and regional temperature change using an ensemble of observational
estimates: the HadCRUT4 data set. J. Geophys. Res. 117 http://dx.doi.org/
10.1029/2011JD017187.
Muscheler, R., Joos, F., Beer, J., Müller, S.A., Vonmoos, M., Snowball, I., 2007. Solar
activity during the last 1000yr inferred from radionuclide records. Quat. Sci.
Rev. 26, 82e97. http://dx.doi.org/10.1016/j.quascirev.2006.07.012.
gin, C., Gennaretti, F., Arseneault, D., Marion, J.,
Naulier, M., Savard, M.M., Be
gin, Y., 2015. A millennial summer temperature reconstruction for
Nicault, A., Be
northeastern Canada using oxygen isotopes in subfossil trees. Clim. Past 11,
1153e1164. http://dx.doi.org/10.5194/cp-11-1153-2015.
Ogurtsov, M., Veretenenko, S., Lindholm, M., Jalkanen, R., 2016. Possible solarclimate imprint in temperature proxies from the middle and high latitudes of
North America. Adv. Space Res. 57, 1112e1117. http://dx.doi.org/10.1016/
j.asr.2015.12.026.
Oman, L., Robock, A., Stenchikov, G., Schmidt, G.A., Ruedy, R., 2005. Climatic
response to high-latitude volcanic eruptions. J. Geophys. Res. 110 http://
dx.doi.org/10.1029/2004JD005487.
Ortega, P., Lehner, F., Swingedouw, D., Masson-Delmotte, V., Raible, C.C., Casado, M.,
Yiou, P., 2015. A model-tested North Atlantic Oscillation reconstruction for the
past millennium. Nature 523, 71e74. http://dx.doi.org/10.1038/nature14518.
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
PAGES 2k-PMIP3 group, 2015. Continental-scale temperature variability in PMIP3
simulations and PAGES 2k regional temperature reconstructions over the past
millennium. Clim. Past 11, 1673e1699. http://dx.doi.org/10.5194/cp-11-16732015.
PAGES2k, 2013. Continental-scale temperature variability during the past two
millennia. Nat. Geosci. 6, 339e346.
Pausata, F.S.R., Chafik, L., Caballero, R., Battisti, D.S., 2015. Impacts of high-latitude
volcanic eruptions on ENSO and AMOC. Proc. Natl. Acad. Sci. 112,
13784e13788. http://dx.doi.org/10.1073/pnas.1509153112.
Payette, S., 2007. Contrasted dynamics of northern Labrador tree lines caused by
climate change and migrational lag. Ecology 88, 770e780.
Pederson, N., Hessl, A.E., Baatarbileg, N., Anchukaitis, K.J., Cosmo, N.D., 2014. Pluvials, droughts, the Mongol Empire, and modern Mongolia. Proc. Natl. Acad. Sci.
111, 4375e4379. http://dx.doi.org/10.1073/pnas.1318677111.
Phipps, S.J., McGregor, H.V., Gergis, J., Gallant, A.J., Neukom, R., Stevenson, S.,
Ackerley, D., Brown, J.R., Fischer, M.J., Van Ommen, T.D., 2013. Paleoclimate
dataemodel comparison and the role of climate forcings over the past 1500
years. J. Clim. 26, 6915e6936.
Plummer, C.T., Curran, M.A.J., van Ommen, T.D., Rasmussen, S.O., Moy, A.D.,
Vance, T.R., Clausen, H.B., Vinther, B.M., Mayewski, P.A., 2012. An independently
dated 2000-yr volcanic record from Law Dome, East Antarctica, including a new
perspective on the dating of the 1450s CE eruption of Kuwae, Vanuatu. Clim.
Past 8, 1929e1940. http://dx.doi.org/10.5194/cp-8-1929-2012.
Pongratz, J., Reick, C., Raddatz, T., Claussen, M., 2008. A reconstruction of global
agricultural areas and land cover for the last millennium. Glob. Biogeochem.
Cycles 22. http://dx.doi.org/10.1029/2007gb003153.
Raspopov, O., Dergachev, V., Esper, J., Kozyreva, O., Frank, D., Ogurtsov, M.,
€m, T., Shao, X., 2008. The influence of the de Vries (~ 200-year) solar
Kolstro
cycle on climate variations: results from the Central Asian Mountains and their
global link. Palaeogeogr. Palaeoclimatol. Palaeoecol. 259, 6e16.
€tzold, J., 2003. Shift in ENSO teleconnections
Rimbu, N., Lohmann, G., Felis, T., Pa
recorded by a northern Red Sea coral. J. Clim. 16, 1414e1422. http://dx.doi.org/
10.1175/1520-0442(2003)16<1414:sietrb>2.0.co;2.
Rind, D., Lean, J., Healy, R., 1999. Simulated time-dependent climate response to
solar radiative forcing since 1600. Geophys. Res. Lett. 104, 1973e1990.
Rind, D., Shindell, D., Perlwitz, J., Lerner, J., Lonergan, P., Lean, J., McLinden, C., 2004.
The relative importance of solar and anthropogenic forcing of climate change
between the Maunder Minimum and the present. J. Clim. 17, 906e929.
Robock, A., 2000. Volcanic eruptions and climate. Rev. Geophys 38, 191e219.
Rutherford, S., Mann, M.E., Delworth, T.L., Stouffer, R.J., 2003. Climate field reconstruction under stationary and nonstationary forcing. J. Clim. 16, 462e479.
Rydval, M., Larsson, L.Å., McGlynn, L., Gunnarson, B.E., Loader, N.J., Young, G.H.,
Wilson, R., 2014. Blue intensity for dendroclimatology: should we have the
blues? Experiments from Scotland. Dendrochronologia 32, 191e204. http://
dx.doi.org/10.1016/j.dendro.2014.04.003.
Rydval, M., Loader, N.J., Gunnarson, B.E., Druckenbrod, D.L., Linderholm, H.W.,
Moreton, S.G., Wood, C.V., Wilson, R., 2017. Reconstructing 800 years of summer
temperatures in Scotland from tree rings. Clim. Dyn. http://dx.doi.org/10.1007/
s00382-016-3478-8.
Schmidt, G.A., 2010. Enhancing the relevance of palaeoclimate model/data comparisons for assessments of future climate change. J. Quat. Sci. 25, 79e87.
Schmidt, G.A., Annan, J.D., Bartlein, P.J., Cook, B.I., Guilyardi, E., Hargreaves, J.C.,
Harrison, S.P., Kageyama, M., LeGrande, A.N., Konecky, B., Lovejoy, S.,
Mann, M.E., Masson-Delmotte, V., Risi, C., Thompson, D., Timmermann, A.,
Tremblay, L.B., Yiou, P., 2014. Using palaeo-climate comparisons to constrain
future projections in CMIP5. Clim. Past 10, 221e250. http://dx.doi.org/10.5194/
cp-10-221-2014.
Schmidt, G.A., Jungclaus, J.H., Ammann, C.M., Bard, E., Braconnot, P., Crowley, T.J.,
Delaygue, G., Joos, F., Krivova, N.A., Muscheler, R., Otto-Bliesner, B.L., Pongratz, J.,
Shindell, D.T., Solanki, S.K., Steinhilber, F., Vieira, L.E.A., 2012. Climate forcing
reconstructions for use in PMIP simulations of the Last Millennium (v1.1).
Geosci. Model Dev. 5, 185e191.
Schneider, D.P., Ammann, C.M., Otto-Bliesner, B.L., Kaufman, D.S., 2009. Climate
response to large, high-latitude and low-latitude volcanic eruptions in the
Community Climate System Model. J. Geophys. Res. 114 http://dx.doi.org/
10.1029/2008JD011222.
Schneider, L., Smerdon, J.E., Büntgen, U., Wilson, R.J.S., Myglan, V.S., Kirdyanov, A.V.,
Esper, J., 2015. Revising midlatitude summer temperatures back to A.D. 600
based on a wood density network. Geophys. Res. Lett. 42, 4556e4562. http://
dx.doi.org/10.1002/2015gl063956.
Schneider, T., 2001. Analysis of incomplete climate data: Estimation of mean values
and covariance matrices and imputation of missing values. J. Clim. 14, 853e871.
Schurer, A.P., Hegerl, G.C., Mann, M.E., Tett, S.F., Phipps, S.J., 2013. Separating forced
from chaotic climate variability over the past millennium. J. Clim. 26,
6954e6973.
Schurer, A.P., Tett, S.F.B., Hegerl, G.C., 2014. Small influence of solar variability on
climate over the past millennium. Nat. Geosci. 7, 104e108. http://dx.doi.org/
10.1038/ngeo2040.
Schweingruber, F., Briffa, K., Nogler, P., 1993. A tree-ring densitometric transect from
Alaska to Labrador. Int. J. Biometeorol. 37, 151e169.
€ker, O., Drew, L., Sch€
Schweingruber, F., Fritts, H., Bra
ar, E., 1978. The X-ray technique
as applied to dendroclimatology. Tree-Ring Bull. 38, 61e91.
Seager, R., Burgman, R., Kushnir, Y., Clement, A., Cook, E., Naik, N., Miller, J., 2008.
Tropical Pacific forcing of North American medieval megadroughts: testing the
concept with an atmosphere model forced by coral-reconstructed SSTs. J. Clim.
21
21, 6175e6190.
Seager, R., Graham, N., Herweijer, C., Gordon, A.L., Kushnir, Y., Cook, E.R., 2007.
Blueprints for medieval hydroclimate. Quat. Sci. Rev. 26, 2322e2336.
Shindell, D., Faluvegi, G., 2009. Climate response to regional radiative forcing during
the twentieth century. Nat. Geosci. 2, 294e300.
Shindell, D.T., 2014. Inhomogeneous forcing and transient climate sensitivity. Nat.
Clim. Change 4, 274e277.
Shindell, D.T., Faluvegi, G., Rotstayn, L., Milly, G., 2015. Spatial patterns of radiative
forcing and surface temperature response. J. Geophys. Res. Atmos. 120,
5385e5403.
Shindell, D.T., Schmidt, G.A., Mann, M.E., Faluvegi, G., 2004. Dynamic winter climate
response to large tropical volcanic eruptions since 1600. J. Geophys. Res. Atmos. 109 http://dx.doi.org/10.1029/2003JD004151.
Shindell, D.T., Schmidt, G.A., Mann, M.E., Rind, D., Waple, A., 2001a. Solar forcing of
regional climate change during the Maunder Minimum. Science 294,
2149e2152.
Shindell, D.T., Schmidt, G.A., Miller, R.L., Mann, M.E., 2003. Volcanic and solar
forcing of climate change during the preindustrial era. J. Clim. 16, 4094e4107.
Shindell, D.T., Schmidt, G.A., Miller, R.L., Rind, D., 2001b. Northern Hemisphere
winter climate response to greenhouse gas, ozone, solar, and volcanic forcing.
J. Geophys. Res. - Atmos. 106, 7193e7210.
Sigl, M., Winstrup, M., McConnell, J.R., Welten, K.C., Plunkett, G., Ludlow, F.,
Büntgen, U., Caffee, M., Chellman, N., Dahl-Jensen, D., Fischer, H., Kipfstuhl, S.,
Kostick, C., Maselli, O.J., Mekhaldi, F., Mulvaney, R., Muscheler, R., Pasteris, D.R.,
Pilcher, J.R., Salzer, M., Schüpbach, S., Steffensen, J.P., Vinther, B.M.,
Woodruff, T.E., 2015. Timing and climate forcing of volcanic eruptions for the
past 2,500 years. Nature 523, 543e549. http://dx.doi.org/10.1038/nature14565.
Smerdon, J.E., Kaplan, A., Amrhein, D.E., 2010. Erroneous model field representations in multiple pseudoproxy studies: corrections and implications. J. Clim. 23,
5548e5554. http://dx.doi.org/10.1175/2010JCLI3742.1.
Smerdon, J.E., Kaplan, A., Chang, D., 2008. On the origin of the standardization
sensitivity in RegEM climate field reconstructions. J. Clim. 21, 6710e6723.
lez-Rouco, J.F., Evans, M.N., 2011. Spatial
Smerdon, J.E., Kaplan, A., Zorita, E., Gonza
performance of four climate field reconstruction methods targeting the Common Era. Geophys. Res. Lett. 38 http://dx.doi.org/10.1029/2011GL047372.
Smerdon, J.E., Pollack, H.N., 2016. Reconstructing Earth's surface temperature over
the past 2000 years: the science behind the headlines. Wiley Interdiscip. Rev.
Clim. Change. http://dx.doi.org/10.1002/wcc.418.
Solomon, A., Goddard, L., Kumar, A., Carton, J., Deser, C., Fukumori, I., Greene, A.,
Hegerl, G., Kirtman, B., Kushnir, Y.M.N., Smith, D., Vimont, D., Delworth, T.,
Meehl, G.A., Stockdale, T., 2011. Distinguishing the roles of natural and
anthropogenically forced decadal climate variability. Bull. Am. Meteorol. Soc.
92, 141e156.
St. George, S., 2014. An overview of tree-ring width records across the Northern
Hemisphere. Quat. Sci. Rev. 95, 132e150. http://dx.doi.org/10.1016/
j.quascirev.2014.04.029.
St. George, S., Ault, T.R., 2014. The imprint of climate within Northern Hemisphere
trees. Quat. Sci. Rev. 89, 1e4.
St. George, S., Ault, T.R., Torbenson, M.C.A., 2013. The rarity of absent growth rings in
Northern Hemisphere forests outside the American Southwest. Geophys. Res.
Lett. 40, 3727e3731. http://dx.doi.org/10.1002/grl.50743.
Steiger, N.J., Hakim, G.J., Steig, E.J., Battisti, D.S., Roe, G.H., 2014. Assimilation of timeaveraged pseudoproxies for climate reconstruction. J. Clim. 27, 426e441. http://
dx.doi.org/10.1175/jcli-d-12-00693.1.
€hlich, C., 2009. Total solar irradiance during the Holocene.
Steinhilber, F., Beer, J., Fro
Geophys. Res. Lett. 36 http://dx.doi.org/10.1029/2009gl040142.
Stenchikov, G., Hamilton, K., Stouffer, R.J., Robock, A., Ramaswamy, V., Santer, B.,
Graf, H.F., 2006. Arctic Oscillation response to volcanic eruptions in the IPCC
AR4 climate models. J. Geophys. Res. 111, D07107. http://dx.doi.org/10.1029/
2005JD006286.
Stoffel, M., Khodri, M., Corona, C., Guillet, S., Poulain, V., Bekki, S., Guiot, J.,
Luckman, B.H., Oppenheimer, C., Lebas, N., Beniston, M., Masson-Delmotte, V.,
2015. Estimates of volcanic-induced cooling in the Northern Hemisphere over
the past 1,500 years. Nat. Geosci. 8, 784e788. http://dx.doi.org/10.1038/
ngeo2526.
von Storch, H., Zorita, E., Jones, J.M., Dimitriev, Y., Gonzalez-Rouco, F., Tett, S.F.B.,
2004. Reconstructing past climate from noisy data. Science 306, 679e682.
Stott, P.A., Gillett, N.P., Hegerl, G.C., Karoly, D.J., Stone, D.A., Zhang, X., Zwiers, F.,
2010. Detection and attribution of climate change: a regional perspective. Wiley
Interdiscip. Rev. Clim. Change 1, 192e211.
Stott, P.A., Jones, G.S., 2009. Variability of high latitude amplification of anthropogenic warming. Geophys. Res. Lett. 36 http://dx.doi.org/10.1029/2009gl037698.
Stott, P.A., Tett, S.F.B., 1998. Scale-dependent detection of climate change. J. Clim. 11,
3282e3294.
http://dx.doi.org/10.1175/1520-0442(1998)011<3282:
sddocc>2.0.co;2.
Stuiver, M., Braziunas, T.F., 1993. Sun, ocean, climate and atmospheric 14CO2: an
evaluation of causal and spectral relationships. Holocene 3, 289e305.
Sundqvist, H.S., Kaufman, D.S., McKay, N.P., Balascio, N.L., Briner, J.P., Cwynar, L.C.,
Sejrup, H.P., Sepp€
a, H., Subetto, D.A., Andrews, J.T., Axford, Y., Bakke, J.,
Birks, H.J.B., Brooks, S.J., de Vernal, A., Jennings, A.E., Ljungqvist, F.C.,
Rühland, K.M., Saenger, C., Smol, J.P., Viau, A.E., 2014. Arctic Holocene proxy
climate database: new approaches to assessing geochronological accuracy and
encoding climate variables. Clim. Past 10, 1605e1631. http://dx.doi.org/10.5194/
cp-10-1605-2014.
lia, D., Servonnat, J.,
Swingedouw, D., Terray, L., Cassou, C., Voldoire, A., Salas-Me
22
K.J. Anchukaitis et al. / Quaternary Science Reviews 163 (2017) 1e22
2010. Natural forcing of climate during the last millennium: fingerprint of solar
variability. Clim. Dyn. 36, 1349e1364. http://dx.doi.org/10.1007/s00382-0100803-5.
Szeicz, J.M., MacDonald, G.M., 1995. Recent white spruce dynamics at the subarctic
alpine treeline of north-western Canada. J. Ecol. 83, 873e885. http://dx.doi.org/
10.2307/2261424.
Tierney, J.E., Abram, N.J., Anchukaitis, K.J., Evans, M.N., Giry, C., Kilbourne, K.H.,
Saenger, C.P., Wu, H.C., Zinke, J., 2015. Tropical sea surface temperatures for the
past four centuries reconstructed from coral archives. Paleoceanography 30,
226e252. http://dx.doi.org/10.1002/2014pa002717.
Tingley, M., Craigmile, P.F., Haran, M., Li, B., Mannshardt, E., Rajaratnam, B., 2015. On
discriminating between GCM forcing configurations using Bayesian reconstructions of late-holocene temperatures. J. Clim. 28, 8264e8281. http://
dx.doi.org/10.1175/jcli-d-15-0208.1.
Tingley, M.P., Craigmile, P.F., Haran, M., Li, B., Mannshardt, E., Rajaratnam, B., 2012.
Piecing together the past: statistical insights into paleoclimatic reconstructions.
Quat. Sci. Rev. 35, 1e22.
Tingley, M.P., Huybers, P., 2010. A Bayesian algorithm for reconstructing climate
anomalies in space and time. part i: development and applications to paleoclimate reconstruction problems. J. Clim. 23, 2759e2781. http://dx.doi.org/
10.1175/2009JCLI3015.1.
Tingley, M.P., Huybers, P., 2013. Recent temperature extremes at high northern
latitudes unprecedented in the past 600 years. Nature 496, 201e205. http://
dx.doi.org/10.1038/nature11969.
Torrence, C., Compo, G., 1998. A practical guide to wavelets. Bull. Am. Meteorol. Soc.
79, 61e78.
Touchan, R., Anchukaitis, K.J., Shishov, V.V., Sivrikaya, F., Attieh, J., Ketmen, M.,
Stephan, J., Mitsopoulos, I., Christou, A., Meko, D.M., 2014. Spatial Patterns of
eastern Mediterranean climate influence on tree growth. Holocene 24,
381e392. http://dx.doi.org/10.1177/0959683613518594.
Trenberth, K.E., 1984. Some effects of finite sample size and persistence on meteorological statistics. part I: autocorrelations. Mon. Wea. Rev. 112, 2359e2368.
http://dx.doi.org/10.1175/1520-0493(1984)112<2359:seofss>2.0.co;2.
Verosub, K.L., Lippman, J., 2008. Global impacts of the 1600 eruption of Peru's
Huaynaputina volcano. Eos. Trans. Am. Geophys. Union 89, 141e142.
Vieira, L.E.A., Solanki, S.K., 2010. Evolution of the solar magnetic flux on time scales
of years to millenia. Astron. Astrophy. 509, A100. http://dx.doi.org/10.1051/
0004-6361/200913276.
Wagner, G., Beer, J., Masarik, J., Muscheler, R., Kubik, P.W., Mende, W., Laj, C.,
Raisbeck, G.M., Yiou, F., 2001. Presence of the solar de vries cycle (~ 205 years)
during the last ice age. Geophys. Res. Lett. 28, 303e306. http://dx.doi.org/
10.1029/2000gl006116.
Wagner, S., Zorita, E., 2005. The influence of volcanic, solar and CO2 forcing on the
temperatures in the Dalton Minimum (1790e1830): a model study. Clim. Dyn.
25, 205e218.
Wang, J., Emile-Geay, J., Guillot, D., McKay, N.P., Rajaratnam, B., 2015. Fragility of
reconstructed temperature patterns over the Common Era: implications for
model evaluation. Geophys. Res. Lett. 42, 7162e7170.
Waple, A., Mann, M.E., Bradley, R.S., 2002. Long-term patterns of solar irradiance
forcing in model experiments and proxy-based surface temperature reconstructions. Clim. Dynam 18, 563e578.
Wiles, G.C., D'Arrigo, R.D., Barclay, D., Wilson, R.S., Jarvis, S.K., Vargo, L., Frank, D.,
2014. Surface air temperature variability reconstructed with tree rings for the
Gulf of Alaska over the past 1200 years. Holocene 24, 198e208. http://
dx.doi.org/10.1177/0959683613516815.
Wilson, R., Anchukaitis, K., Briffa, K.R., Büntgen, U., Cook, E., D'Arrigo, R., Davi, N.,
Esper, J., Frank, D., Gunnarson, B., Hegerl, G., Helama, S., Klesse, S., Krusic, P.J.,
Linderholm, H.W., Myglan, V., Osborn, T.J., Rydval, M., Schneider, L., Schurer, A.,
Wiles, G., Zhang, P., Zorita, E., 2016. Last millennium northern hemisphere
summer temperatures from tree rings: Part I: the long term context. Quat. Sci.
Rev. 134, 1e18. http://dx.doi.org/10.1016/j.quascirev.2015.12.005.
Wilson, R., Cook, E., D'Arrigo, R., Riedwyl, N., Evans, M.N., Tudhope, A., Allan, R.,
2010. Reconstructing ENSO: the influence of method, proxy data, climate
forcing and teleconnections. J. Quat. Sci. 25, 62e78. http://dx.doi.org/10.1002/
jqs.1297.
Wilson, R., D'Arrigo, R.D., Buckley, B.M., Büntgen, U., Esper, J., Frank, D., Luckman, B.,
Payette, S., Vose, R., Youngblut, D., 2007. A matter of divergence: tracking recent
warming at hemispheric scales using tree ring data. J. Geophys. Res. 112 http://
dx.doi.org/10.1029/2006JD008318.
Wilson, R., Rao, R., Rydval, M., Wood, C., Larsson, L.A., Luckman, B.H., 2014. Blue
Intensity for dendroclimatology: the BC blues: a case study from British
Columbia, Canada. Holocene 24, 1428e1438. http://dx.doi.org/10.1177/
0959683614544051.
Wilson, R., Tudhope, A., Brohan, P., Briffa, K.R., Osborn, T.J., Tett, S., 2006. Twohundred-fifty years of reconstructed and modeled tropical temperatures.
J. Geophys. Res. 111, C10007. http://dx.doi.org/10.1029/2005JC003188.
Wise, E.K., 2015. Tropical pacific and northern hemisphere influences on the
coherence of pacific decadal oscillation reconstructions. Int. J. Climatol. 35,
154e160.
Youngblut, D., Luckman, B., 2008. Maximum June-July temperatures in the southwest Yukon over the last 300 years reconstructed from tree rings. Dendrochronologia 25, 153e166. http://dx.doi.org/10.1016/j.dendro.2006.11.004.
Zanchettin, D., Timmreck, C., Graf, H.F., Rubino, A., Lorenz, S., Lohmann, K.,
Krüger, K., Jungclaus, J.H., 2012. Bi-decadal variability excited in the coupled
oceaneatmosphere system by strong tropical volcanic eruptions. Clim. Dyn. 39,
419e444. http://dx.doi.org/10.1007/s00382-011-1167-1.
€rklund, J., Chen, D., 2016. 1200
Zhang, P., Linderholm, H.W., Gunnarson, B.E., Bjo
years of warm-season temperature variability in central Scandinavia inferred
from tree-ring density. Clim. Past 12, 1297e1312. http://dx.doi.org/10.5194/cp12-1297-2016.
lez-Rouco, F., Legutke, S., 2003. Testing the Mann et al. [1998]
Zorita, E., Gonza
approach to paleoclimate reconstructions in the context of a 1000-yr control
simulation with the ECHO-G Coupled Climate Model. J. Clim. 16, 1378e1390.
Zwiers, F.W., Weaver, A.J., 2000. The causes of 20th century warming. Science 290,
2081e2083.